Studentnr: s2402718 Name: Stefan Konterman Study Program: MSc Finance Supervisor: Dr. A. Plantinga
The effect of perceived and actual financial literacy on the risk attitude
of personal investors.
Abstract
2 Table of Contents 0. Introduction ... 3 1. Literature review ... 6 1.1 Financial literacy ... 6 1.2 Risk attitude ... 9 1.3 Hypotheses ... 10 2. Methodology ... 11 3. Data ... 13 3.1 Financial literacy ... 13 3.2 Risk attitude ... 19 4. Results ... 21
4.1 Actual risk attitude ... 22
4.2 Perceived risk attitude ... 25
5. Conclusion ... 29
References ... 31
3
0. Introduction
Risk is an inherent feature of all investment options even bank savings books can contain risks, as some discovered in the course of the financial crisis in 2008. When asked most people answer what risk means by saying that they recall images of dangers in everyday surroundings like traffic. The term risk attitude used in the scientific publications is generally assumed to represent the willingness of people to take a certain amount risk to reach their objectives. For instance, stocks are regarded as risk full investments. Risk attitude is an important variable for personal investors to decide between different investments. Every personal investor considers the risk before proceeding into financial investments. Required is that personal investors have knowledge of many financial products to make a deliberate decision on future investments. The different options in financial products for short and long term investments are huge, while individual product details may differ in subtle ways with large impacts for future outcomes. The results can vary significantly when the wrong consideration is made in which to invest. Financial literacy is indispensable. The
interpretation of financial literacy is not bounded by one definition. Some of the definitions for financial literacy used in research are: knowledge of financial concepts, ability to communicate about financial concepts, aptitude in managing personal finances and skill in making appropriate financial decisions or confidence to plan effectively for future financial needs (Remund 2010). Not only on financial markets knowledge is necessary, for households on everyday decisions financial expertise is required. Knowledge of insurance, savings, credit cards, mortgages, future investments and taxes needs to be present to take on well-informed decisions.
Microeconomic models (Modigliani and Brumberg 1954; Friedman 1957) from the early financial era posits that individual investors who behave rational will consume less than their full income during times of high earnings. A rational consumer is expected to arrange his savings and income to spread marginal utility over his lifetime. Multiple studies have shown that such optimizations can be shaped by consumer preferences (Browning and Lusardi 1996; Attanasio and Weber 2010). These models assume that individuals have the financial
4
The decision to invest in financial products depends on multiple variables, among which is risk attitude. Personal investors have to make a well-considered decision if the investment is worth the risk. To understand the risks involved with investments an
understanding of the consequences is necessary. Education of personal finance is crucial for personal investors to reach the necessary level of financial literacy. Although not every consumer is benefitted by education. Due to lack of time or interest most consumers do not reach the appropriate level of financial literacy. Recently Willis, L. (2008) studied the effect of financial education on financial literacy of consumers. He finds in his study that it is widely believed that financial-literacy education is necessary for consumers to understand the more complex financial products. The financial education is widely believed to turn consumers into ‘responsible’ and ‘empowered’ market players, motivated and competent to make financial decisions that increase their own welfare. Although this vision is seductive, promising both a free market and increased consumer welfare, the predicate belief in the effectiveness of financial-literacy education lacks empirical support. He finds that for some consumers
financial education appears to increase confidence without improving ability, leading to worse financial decisions. People tend to overestimate their knowledge of financial products, and therefore have a higher perceived financial literacy.
These consumers may not reach an actual higher level of financial literacy, they do contemplate a higher confidence in their own abilities. Financial education, if applied not consequently and keeping up-to-date will not increase actual financial literacy but can cause an increase in confidence, perceived knowledge (Willis, L. 2008). Investors can be
overconfident about their own understanding of the risks for investing. Higher confidence in own abilities raises the perceived financial literacy of consumers, the consumers think they know more than is actually true. Overconfidence can be the cause of investments going wrong when people value their own knowledge higher than is factual. The perceived literacy is not in line with the actual literacy (Ellen, J., 1975, Barber, B., Odean, T., 2001). If investors do not have the full grasp of financial products, he cannot make an adequate decision on whether the risk is acceptable. A recent example of misunderstood risk is the Legiolease affair (Dexia Bank Nederland), which concerned the lease of securities. A financial product was sold to low-income consumers by lending money to them with which they could ‘lease’ stock. Stock investments were made available to those who could otherwise not invest in financial
products. The consumers were left in the assumption that the monthly payments were
5
markets went down, a lot of investors were charged with large amounts of debt. If investors were financial literate, could the Legiolease affair have been prevented? Are literate investors more reserved about taking risks? One of the complaints consumers of the Legiolease affair were that the risks were not explained to them extensively enough.
The aim of this thesis is to obtain insight in the relationship between financial literacy and risk attitude. In presenting a clear result a distinction is made between actual and
perceived financial literacy, both can affect financial behavior (Allgood and Walstad, 2012). The risk perception is affected by literacy as Sachse, Jungermann and Belting, 2011, find in their study. In this study a distinction is made between actual risk attitude and perceived risk attitude. To achieve this goal the analysis is based on a representative survey of the population of the Netherlands. The research question in this thesis will be:
What is the effect of financial literacy on the risk attitude of personal investors?
The thesis is structured as follows: The first chapter gives an overview of the existing literature. The definition of financial literacy is introduced and various researches are
6
1. Literature review
1.1 Financial literacy
When conducting research on financial literacy a significant challenge is to determine how best to measure financial literacy, there is no standard definition in the research literature (Hung et al. 2009, Huston 2010, Remund 2010). No standard definition in research literature is available, the label for perceived and actual financial literacy is not consistent either. Actual financial literacy is addressed in literature as financial knowledge and financial literacy. Perceived financial literacy is addressed as confidence, perceived knowledge and perceived financial literacy (Lusardi and Mitchell, 2007b, Van Rooij, Lusardi and Alessie, 2007, Parker, Yoong, Bruine de Bruin, and Willis, 2011, Bruine de bruin, 2007). Most research on financial literacy is conducted using multiple choice test questions and/or true-false test questions that are embedded in questionnaires which contains questions about demographic characteristics and asks about financial behaviors and activities. (Hilgert et al. 2003, Hung et al. 2009, Huston, S, 2010). Questionnaires on financial literacy have been used by economists in explaining financial behavior, such as financial behavior on retirement planning (Lusardi and Mitchell 2007; Lusardi and Mitchell, 2008; van Rooij et al. 2011a; Lusardi and Mitchell 2011), wealth accumulation (Behrman et. al. 2012; Gustman et al. 2012), stock investing (Abreu and Mendes 2010; van Rooij et al. 2011b) and inflation expectations (Bruine de Bruin et al. 2010).
An oversight of multiple researches on financial literacy is presented below. Consistent through all research is the study of relationship between financial literacy and financial behavior.
7
score high on the actual financial literacy test. Their study show a high positive correlation between perceived financial literacy and actual financial literacy. Even after controlling for a range of socioeconomic factors, financial literacy still has its own independent effect.
Although education boosts the probability of retirement planning.
Van Rooij, Lusardi and Alessie, 2007, devised two special modules to apply in the Dutch Household Survey (DHS) for measuring financial literacy and its relationship to the stock market participation. The two literacy modules were added to the survey in 2005 and 2006. To measure and evaluate financial literacy the financial literacy questions are composed of two parts. The first set of questions aims to assess basic financial literacy and the second set aims to assess advanced financial literacy. The combined set of question gives valid
representation of the actual financial literacy of participating respondents. The modules were designed using similar modules as in the HRS and a variety of other surveys on financial literacy. In this study a few questions are unique to measure financial literacy (Lusardi and Mitchell, 2011, Lusardi and Mitchell, 2007b). The literacy modules contain an objective and a subjective way of measuring literacy. The module include a question in which respondents are asked to self-asses their financial knowledge. The subjective way of measuring financial literacy is labeled in literature as perceived financial literacy (Agnew and Szykman, 2005, Hung et al., 2009, Lusardi and Mitchell, 2011). To control for individual characteristics, control variables are included in the model. Gender, education, income and wealth are shown important predictors of participation in the stock market. Although after controlling for demographic characteristics, financial literacy also matters for stock ownership. The different measures of financial knowledge show that lack of literacy prevents households from
participating in the stock market. On the other hand it is not clear that unsophisticated, low financial literate, investors would be able to take full advantage of the stock market and invest efficiently. Low literate personal investors are more likely to make unwise choices when investing in financial markets and take on higher risks. Financial decision-making is affected by financial literacy.
Parker, Yoong, Bruine de Bruin, and Willis, 2011, find in their research that
confidence in knowledge predicts self-reported retirement planning and savings, as well as performance on a hypothetical investment task, independently of the effect of actual
8
interaction between perceived and actual financial literacy is worth studying for explaining the wide range of individual differences. Individuals may show high perceived literacy but a low level of actual literacy. The opposite can occur evenly. Also both high perceived and high actual literacy may occur. In their study Parker, Yoong, Bruine de Bruin, and Willis, 2011, examined the relationship of confidence with self-reported and actual financial decisions, using four different tasks, each performed by overlapping samples of ALP participants. The four studies were designed by different researchers, using different domains and
methodologies, thus allowing to examine the robustness of results across these approaches. Study 1 designed by Lusardi and Mitchell, 2009, includes thirteen true/false questions of financial knowledge, with overall confidence assessed with a single question assessing perceived knowledge. Study 2 designed by Parker, Bruine de bruin and Fischhoff, 2007, included a form of the under/overconfidence instrument designed for the Adult Decision-Making Competence measure. It consists of fourteen true/false questions measuring general knowledge, with confidence in each answer being assessed on a scale ranging from 50% to 100%. Study 3 is designed by Dominitz, Hung and Yoong, 2009, and included a hypothetical fee minimization when investing task which asked respondents to allocate a fixed amount of money among index funds. Overall confidence was assessed with a single question asked on a scale from 1 to 5. The final study was conducted by Delavande, Rohwedder and Willis, 2008, Kimball, 2008, Willis, 2008, and included 70 true/false questions measuring financial
sophistication with responses given on a 12-point scale. After presenting parallel analysis across the four separate studies the main finding is that the confidence is positively correlation with knowledge. In this study it is concluded that individuals who experience overconfidence, high perceived knowledge, are more likely to carefully read through investment options or make better use of financial planners.
Parker, Bruine de bruin and Fischhoff, 2007, studied the differences in levels of confidence, perceived knowledge, opposed to actual knowledge. He finds in his research that individuals with inappropriate levels of confidence, as seen in differences in expressed percent confidence and the percent of accurate responses across knowledge items, are more likely to exhibit framing errors and to violate the rules of probability when judging risks. Individuals who overestimate their own knowledge opposite to their actual knowledge tend to judge risks lower or more controllable.
9
literacy. They find that the combined measure provides greater understanding about how financial literacy affects financial behaviors. A large national survey of U.S. adults and households was used to investigate how this overall financial literacy affects financial behaviors across five financial topics. To control for individual characteristics, control variables are constructed. The list of control variables includes gender, age, race, educations, marital status, employment, income. The analysis show that financial literacy is the most consistent and often the most influential factor, and that the demographic factors are of secondary importance. The results from the probit analysis show that both actual and perceived financial literacy significantly influence financial behaviors.
Consistent conclusions from the different research conducted in financial literacy can be stated as that financial literacy have a significant effect on financial decision-making. The correlation between actual financial literacy and perceived financial literacy is in most
researches positive. Which concludes that when respondents perceive their own knowledge as high, their actual knowledge is high too.
1.2 Risk attitude
In the Oxford English Dictionary risk is described as a situation involving exposure to danger or from a financial view the possibility of a financial loss. Olsen, 1997, investigated in his study the perception of investment risk among participants and concludes that loss of capital, returns below expectation and economic uncertainty are prevalent associations with risk. Other responses are perceived knowledge deficits and a feeling of lack of control.
Weber and Milliman, 1997, Study the relation between risk perception and risky choices. In the paper the risk-return framework is examined. Among the characteristics of the risk-return framework is that an investor’s preference or attitude towards risk is assumed to determine his or her risk-value tradeoff. When individual investors are invested in stock, their risk-return tradeoff is perceived as high. Concluding from these statements is that consumers who are invested in stocks are risk seeking or risk loving. Evidence is reviewed that people’s perceptions of the riskiness of choice alternatives are not always captured by conventional risk indices and can differ significantly from individual to individual.
10
specificity of risk perception of various common investment products. The analysis showed that self-assessed knowledge is not correlated with perceived investment risk. The second study focuses on effects of individual characteristics on financial risk perception. Only
financial literacy proves to be relevant in a regression analysis were perceived investment risk is explained by using gender, age, investment experience and financial literacy as predictors.
Diacon, 2004, presents detailed results of the comparison of perceptions by individual consumers and expert financial advisers of the investment risk involved in various UK personal financial services’ products. Tests show that there are significant differences between expert and lay investors in the way financial risks are perceived. Financial experts are likely to be less loss averse than lay investors, but are prone to affiliation bias, believe that the products are less complex, and are less cynical and distrustful about the protection
provided by the regulators. The traditional response to the finding that experts and non-experts have different perceptions and understandings about risk is to re-educate consumers. A common factor which explains variances between experts and consumers is knowledge. The actual financial literacy have to rise in order to change the perception of risk of individual investors.
The literature concludes that the perception of risk differs between levels of financial literacy. Knowledge is one of the common factors which alters the perception of risk. A distinction can be made between actual risk and perceived risk of investments. From studies it shows that perceived knowledge is not correlated with perceived risk, while financial literacy is a significant factor in assessing risk. From the risk-return framework studied by Weber and Milliman, 1997, stock ownership is regarded as a high value-risk tradeoff and thus personal investors involved in stock ownership are assessed as risk-loving.
1.3 Hypotheses
To test whether financial literacy has an effect on the risk attitude of personal investors. The following hypotheses are made.
H0: Financial literacy has no effect on risk attitude.
The main hypothesis is followed up by the following hypothesis to answer the main question. H1: Actual financial literacy has an effect on actual risk attitude.
11
2. Methodology
To test for the effects of financial literacy on the risk attitude of personal investors a model needs to be selected. The dependent variables consist of binary data. The linear probability model (LPM) is by far the simplest way of dealing with binary dependent variables, and it is based on an assumption that the probability of an event occurring is linearly related to a set of explanatory variables. Limitations of the LPM are that probabilities are restricted to the interval zero and one (Maddala 1983, Cox and Snell, 1989). To overcome this limitations a logit model can be estimated. The logit model overcomes the limitations that it can produce estimated probabilities that are negative or greater than one. A logistic function is used that effectively transforms the regression model so that the fitted values are bounded within the zero and one interval. The fitted regression model will visually appear as an S-shape rather than a straight line.
The logit model is specified to investigate effects of financial literacy, overall financial literacy as each type of financial literacy (actual and perceived) on risk attitude of personal investors. The dependent variables are the risk attitude based on stock ownership and the risk attitude based on perceived risk which are listed in the variables table. The set of control variables includes the four financial literacy variables, with the constant as the omitted group. The other variables in the equation are the demographic characteristics and the variables belonging to them. The following variables are included in the equation: gender, age, education and income. Because of the use of dummies for all variables, a selection of variables have to be omitted to prevent for the dummy variable trap. The omitted variables are; female; age <25 year; elementary school; income <4.5K
As stated before logit models are nonlinear regressions where coefficients are fitted with the maximum likelihood to the following function:
𝐹(𝑧𝑖) = 𝑒 𝑧𝑖 1 + 𝑒𝑧𝑖 = 1 1 + 𝑒−𝑧𝑖 𝑧𝑖 = 𝛽1+ 𝛽2𝑥2𝑖+ 𝛽3𝑥3𝑖+ ⋯ + 𝛽𝑘𝑥𝑘𝑖+ 𝑢𝑖
Where zi is a vector of explanatory variables, and βi is vector coefficients to be estimated. The function F is the cumulative logistic distribution. So the logistic model estimate is:
𝑃𝑖 =
1
1 + 𝑒−(𝛽1+ 𝛽2𝑥2𝑖+ 𝛽3𝑥3𝑖+⋯+ 𝛽𝑘𝑥𝑘𝑖+𝑢𝑖)
Where the Pi is the probability that yi = 1. ln ( 𝑝𝑖
1 − 𝑝𝑖
12
With the logistic model, 0 and 1 are asymptotes to the function and thus the probabilities will never actually fall to exactly zero or rise to one. The marginal effect for each dummy
regressor is the change in the log likelihood of the dependent variable equaling one computed for a discrete change in the dummy variable from zero to one when evaluating all other variables at their means. To compute robust standard errors for the logit regression, the robust covariances of Huber/White are used. The model is nonlinear in β which means that
interpreting the logit coefficients is difficult. The marginal effects are therefore presented. The robust z-values are reported next to the marginal effects. In the results the focus is on the interpretation of the marginal effects from comparing the four financial literacy groups; (I) high perceived financial literacy and high actual financial literacy (HPHA); (II) high perceived financial literacy and low actual financial literacy (HPLA); (III) low perceived financial literacy and high actual financial literacy (LPHA); (IV) low perceived financial literacy and low actual financial literacy (LPLA). The most important estimate is the
13
3. Data
The Dutch Household Survey (DHS) is started in 1993 and is commissioned by De
Nederlandsche Bank and conducted by CentERdata. The primary purpose of this study is to study the economic and psychological determinants of the financial behavior of households. The DHS is one of the most extended surveys that takes place which records the financial and economic situation of a representation of the population of the Netherlands.
The survey was completed by administrating questionnaires to 2,000 Dutch households. Every year the set of items on each questionnaires is essentially the same. The CentERdata website provides a copy of each questionnaire, a brief report of survey methods and basic findings, and an SPSS or STATA data file for each sample that can be used by researchers for further analysis. This study uses the data set published in October 2005 and collected from February 2005 t/m August 2005 using an online survey of among 2,277 respondents. The questionnaire administered to each adult was extensive and is divided in seven sections. The first section presents a number of general questions about age, gender, education, marital status, work status, numbers of children in household, area of living and living arrangement. The remaining six sections focus on a wide assortment of financial topics containing;
household and work, living and mortgage, health and income, property and loans, economic and psychological concepts and extra questions related to the DHS. For our study we used the section property and loans and economical and psychological concepts. The results on stock ownership is used and the question that describes the risk perception of the respondents in a scale from 1 to 5, descriptive statistics are shown in table 6.
3.1 Financial literacy
Actual financial literacy is measured using a set of sixteen questions. Van Rooij, Lusardi and Alessie (2011) designed two question sets for measuring the respondent’s financial literacy. One question set relates to basic financial knowledge and the second question set relates to advanced financial knowledge of the households. The first set of five questions aims to assess basic financial literacy. The questions cover topics ranging from the working of interest rates and interest compounding to the effect of inflation, discounting, and nominal versus real values. The questions appear to be relatively simple, although many adults found the
14
27.4 percent. As seen in table 1 while many respondents answer each individual question correctly, the proportion of respondents who answered all five questions correctly is only 44.2 percent. Respondents possess parts of knowledge, but not on all domains. The exact wording of the basic financial questions is presented in appendix I.
TABLE 1: Basic financial literacy.
Panel A reports the proportion of households providing correct and incorrect answers to each of the five basic literacy questions. Panel B reports the distribution of the number of correct and incorrect answers on the five basic literacy questions. The data are from the 2005
DHS. Note: Incorrect answers also include do not know answers. Panel A: Basic financial literacy
Weighted percentages of correct and incorrect answers (N = 1,467)
Numeracy Interest compounding Inflation Time value of money Money Illusion Correct 93.6 80.9 85.6 75.2 72.6 Incorrect 6.4 19.1 14.4 24.8 27.4
Panel B: Summary of responses
Weighted number of correct and incorrect answers (N = 1,467)
Number of correct and incorrect answers (out of five questions)
None 1 2 3 4 All Mean
15
The second set of eleven questions measure more advanced financial knowledge and covers topics such as the difference between stocks and bonds, the function of the stock market, the workings of risk diversification, and the relationship between bond prices and interest rates. The exact wording of the advanced questions are presented in appendix II. These questions measure the ability to perform simple calculations (Van Rooij, Lusardi and Alessie, 2011). The results, presented in table 2, compared to the basic financial questions is much different. The proportion of correct answers on each question is lower. Panel B of table 2 shows that only a very small fraction of total respondents is able to answer all questions correct. Concluded from the result is that financial literacy is limited among the respondents.
Table 2: Advanced financial literacy.
Panel A reports the proportion of households providing correct and incorrect answers to each of the eleven advanced literacy questions. Panel B reports the distribution of the number of correct and incorrect answers on the eleven advanced literacy questions. The data are from the 2005 DHS.
Note: Incorrect answers also include do not know answers. Panel A: Advanced financial literacy
Weighted percentages of correct and incorrect answers (N = 1,467)
6 7 8 9 10 11 12 13 14 15 16
Correct 69.8 65.3 70.9 60.7 26.8 23.8 64.5 51.3 72.2 67.3 33.5
Incorrect 30.2 34.7 29.1 39.3 73.2 76.2 35.5 48.7 27.8 32.7 66.5
Panel B: Summary of responses
Weighted number of correct and incorrect answers (N = 1,467)
Number of correct and incorrect answers (out of eleven questions)
None 6 7 8 9 10 11 12 13 14 15 All Mean
16
To measure the perceived financial literacy the questionnaire contained a self-assessment. Respondents were asked to rate their own knowledge on finance. The question is based on a seven-point scale with a rating of one being very low and seven being very high. The question to self-assess their knowledge gives insight on the perception of the level of financial literacy for individual respondents without having to answer test questions. The results are shown in table 3. The mean is 4.73, showing that most respondents classify themselves as having some financial knowledge.
TABLE 3: Self-assessed financial knowledge.
The results presented in this table show the perception of knowledge for respondents. Labeled as perceived financial literacy. The data is collected from the DHS 2005. Perceived financial literacy
Weighted percentages of value given (N = 1,467)
Question 1 2 3 4 5 6 7 Mean
How would you rate your
own financial knowledge? 0.6 3.8 9.3 24.9 34.0 24.2 3.1 4.73
17
TABLE 4: Scoring financial literacy.
The results in this table show the scoring of respondents. Divided in subcategories. Most people rate themselves as high financial literate, the test on financial literacy show that the
score is lower so there could be misinterpreting their own knowledge (Parker, Bruine de bruin and Fischhoff, 2007).
Scoring respondents on perceived and actual financial literacy (N = 1,467) Perceived financial literacy Actual financial literacy
Low High Low High
568 899 774 696
Control variables are created to control for demographic variables before the relation between financial literacy and risk attitude is assessed (Lusardi and Mitchell, 2007b, van Rooij,
Lusardi, Alessie, 2007, Allgood and Walstad, 2012).To construct these variables information is used from the financial literacy modules composed by van Rooij, Lusardi an Alessie, 2007. Included are variables for age, gender, education, income. For each of these categories
18
TABLE 5: Variable characteristics.
The table shows the descriptive statistics of the variables used to study for relationship between financial literacy and risk attitude. Note: the variable with
asterisks are continuous variables, all other variables are dummies.
Variables N Min Max Mean Std. Deviation
Risk attitude stock ownership 1,115 0 1 0.1453 0.3526 Risk attitude self-assessment 1,037 0 1 0.0482 0.2143 Perceived financial literacy* 1,467 1 7 4.7280 1.1646 Actual financial literacy* 1,467 0 16 10.1759 3.5611
Actual High 1,467 0 1 0.5256 0.4995
Actual Low 1,467 0 1 0.4744 0.4995
Perceived High 1,467 0 1 0.6128 0.4873
Perceived Low 1,467 0 1 0.3872 0.4873
Perceived High/Actual High 1,467 0 1 0.3701 0.4830 Perceived High/Actual Low 1,467 0 1 0.2427 0.4288 Perceived Low/Actual High 1,467 0 1 0.1554 0.3624 Perceived Low/Actual Low 1,467 0 1 0.2318 0.4221
Male 1,467 0 1 0.5556 0.4971 Age <25 1,467 0 1 0.0204 0.1416 Age 26 - 45 1,467 0 1 0.3967 0.4894 Age 46 - 65 1,467 0 1 0.4015 0.4904 Age >65 1,467 0 1 0.1813 0.3854 Elementary 1,467 0 1 0.0443 0.2058 Middle school 1,467 0 1 0.5549 0.4971 High school 1,467 0 1 0.4008 0.4902 Income <2.5K 1,467 0 1 0.8787 0.3266 Income 2.5-4.5K 1,467 0 1 0.1009 0.3013 Income >4.5K 1,467 0 1 0.0068 0.0823
19
survey. The ratio of perceived financial literacy is 0.57 (four divided by seven), while the ratio of actual financial literacy is 0.63 (ten divided by sixteen). Respondents who rate their own financial knowledge as high, most likely score high on the actual financial knowledge test, corresponding to the results of Lusardi and Mitchell, 2007b. The results show that roughly two-third of the respondents consider their own financial knowledge as high, 0.6128. The actual financial literacy shows a similar result, may it be less biased towards high with a result of 0.5256 of the respondents score high in the survey. The results show furthermore that more than half of the respondents are male. Almost all respondents are in age categories 26 - 45 and 46 - 65. Half of the respondents completed middle school as their highest education. The income dummies show that almost 9 out of 10 respondents have an income lower than 2.5K. The financial literacy dummies which are created to measure for literacy show that the high perceived and high actual is the largest group with a result of 0.3701. The dummy
perceived low and actual high is the smallest group with a result of 0.1554. From the literature this corresponds to the results of Lusardi and Mitchell, 2007b, overall respondents who rate their own knowledge as low, score low on actual knowledge. Remarkable result is the group perceived high and actual low, the result show that this group is larger than perceived low and actual low. This is in contrast with results of Lusardi and Mitchell, 2007b. The small sample size of only 1,467 can be a cause of this deviating result.
3.2 Risk attitude
20
TABLE 6: Risk measurement
To measure the risk attitude of respondents who elaborate in the Dutch Household Survey (DHS) the following two questions are considered to show a truthful representation of the willingness to take on investment risk. The first question shows the actual investment in stocks, the latter shows
how respondents perceive the risk in their own decision making.
Question Answers N Min Max Mean Std. Deviation
Did you have investments in stocks on 31 December 2004? Stocks of own
company are not counted for.
0. No
1. Yes 1,115 0 1 0.1453 0.3526
How would you describe the risks that you have taken lately in your investments? When non investments have taken place choose NA.
1. Non risk
2. Some small risk 3. Small risk 4. Some large risk 5. Large risk 6. N.A
7. Don't know
21
4. Results
The available literature predicts that higher financial literacy increases the likelihood of participation in ownership of stocks (van Rooij, Lusardi and Alessie 2007). When using the ownership of stocks as a measurement for risk attitude the same results are to be expected. To test whether the actual financial literacy and perceived financial literacy are measuring the same characteristics a correlation test is performed. As shown in table 7 the correlation between perceived financial literacy and actual financial literacy is only 0.2758, meaning that perceived financial literacy and actual financial literacy is significantly different from each other. This low correlation is consistent with other findings in the research literature (Parker, Bruine de bruin, Yoong and Willis, 2011).
To test whether testing for actual risk attitude and perceived risk attitude are not simply measuring the same characteristics a correlation test is conducted. In table 8 the results are shown. The correlation between perceived risk attitude and actual risk attitude is only 0.2570. The correlation statistic show that the dependent variables have different characteristics.
TABLE 8: Correlation risk attitude.
Test for correlation between actual risk attitude and perceived risk attitude.
Perceived risk attitude
Actual risk attitude 0.2570
TABLE 7: Correlation literacy.
Test for correlation between perceived financial literacy and actual financial literacy to test whether the variables are not the same. The continuous variables as presented in table 5
perceived knowledge and actual knowledge are used to test.
Perceived financial literacy
22 4.1 Actual risk attitude
As shown in table 9 financial literacy is a significant and major influence in explaining the probability for ownership of stocks, high risk attitude.
TABLE 9: Regression output actual risk attitude.
Panel A shows the regression output with actual risk attitude as the dependent variable. Panel B presents the Wald test conducted to test whether the coefficients were equal. Panel A
Coefficients Mean Marginal effects z-Statistic
HPHA (I) 0.9088 0.3701 0.0820 2.9337 HPLA (II) -0.9450 0.2427 -0.0852 -2.8874 LPHA (III) -0.0891 0.1554 -0.0080 -0.2281 LPLA (IV) -1.4212 0.2318 -0.1282 -2.8893 Male 0.1897 0.5556 0.0171 0.8887 Age 26-45 -0.0620 0.3967 -0.0056 -0.1274 Age 46-65 0.0626 0.4015 0.0056 0.1305 Age 65< 0.5008 0.1813 0.0452 1.0073 Income <2.5K -1.6122 0.8787 -0.1454 -3.0785 Income 2.5-4.5 -0.8916 0.1009 -0.0804 -1.5333 Middle school -1.0214 0.5549 -0.0921 -2.7465 High school -0.4959 0.4008 -0.0447 -1.3101
Wald test coefficients Panel B
Null Hypothesis Chi-squared statistic P-value
I = II 35.59 0.00 I = III 13.2 0.00 I = IV 28.63 0.00 II = III 5.16 0.02 II = IV 0.95 0.33 III = IV 6.93 0.01
23
market, and thus are taking more risks. The respondents categorized in the group HPLA (II) are likely to be 8.5% less invested in the stock market, and their risk attitude is biased to risk aversion. They do not take on risks on the stock market. The same is concluded for the LPLA (IV) classification, these respondents are 12.8% less likely to be invested in the stock market. The classification LPHA (III) shows a different pattern, the marginal effect for this group is only negative 0.8%. Respondents who perceive their own knowledge lower as the actual knowledge are little less likely to be invested in the stock market. The z-statistic for group LPHA (III) is not significant, while for the other groups the z-statistic shows significance at the two-tailed 5% significance level. Concluded is that HPHA (I), HPLA (II) and LPLA (IV) are strong predictors of stock ownership. The marginal effect conclude that respondents classified as HPHA (I) have a more risky attitude than the respondents with other
characteristics. The effect of demographic variables across the equations are less consistent. For example, males relative to females are insignificant. They do not predict the ownership of stocks. The age variables show the same results and are not significant as opposed to the omitted variable of lower than 25 years. Income variables are not consistent either, income levels below 2.5K show a significant effect on ownership of stocks measured against income levels above 4.5K. Income between 2.5K and 4.5K do not show a significant effect on stock ownership of respondents. The marginal effect for the income level below 2.5K is negative 14.5%. Respondents with income below 2.5K are 14.5% less likely to be invested in stocks opposed to respondents with income above 4.5K. Education appears to influence stock ownership, but not always. The omitted group for this variable is elementary school. Relative to this group respondents who completed middle school are 9.2% less likely to be invested in stocks. Completion of high school has no significant effect on the ownership of stocks. A Wald test is performed to test that the marginal effects were equal. Table 9 shows that for all literacy variables the coefficients are statistically different except for HPLA (II) - LPLA (IV).
24
TABLE 10: Robustness check actual risk attitude.
Panel A shows the results of a robustness check using half of the dataset to check for consistency across the variables. Panel B shows the linear probability model (LPM) results
Panel A
Complete dataset (N = 1,115) Half dataset (N = 738) Marginal effects z-Statistic Marginal effects z-Statistic
HPHA (I) 0.0820 2.9337 0.0705 2.1377 HPLA (II) -0.0852 -2.8874 -0.0679 -1.7733 LPHA (III) -0.0080 -0.2281 -0.0525 -1.1186 LPLA (IV) -0.1282 -2.8893 -0.0918 -1.7293 Male 0.0171 0.8887 -0.0125 -0.5377 Age 26-45 -0.0056 -0.1274 0.0276 0.6207 Age 46-65 0.0056 0.1305 0.0257 0.5741 Age 65< 0.0452 1.0073 0.0931 1.9862 Income <2.5K -0.1454 -3.0785 -0.1054 -2.6397 Income 2.5-4.5 -0.0804 -1.5333 -0.0235 -0.5963 Middle school -0.0921 -2.7465 -0.1409 -2.8293 High school -0.0447 -1.3101 -0.0663 -1.1030 Panel B
Linear probability model
25 4.2 Perceived risk attitude
Table 11 shows the outcomes of the regression with perceived risk attitude as the dependent variable. Results present that respondents who are classified in the HPHA (I) are 3.5% more likely to perceive their actions as risky. However the results show that for all literacy
26
TABLE 11: Regression output perceived risk attitude.
Panel A shows the regression output with perceived risk attitude as the dependent variable. Panel B presents the Wald test conducted to test whether the coefficients were equal. Panel A
Coefficients Mean Marginal effects z-Statistic
HPHA (I) 0.7946 0.3701 0.0351 1.7230 HPLA (II) -0.6252 0.2427 -0.0276 -1.2155 LPHA (III) 0.5125 0.1554 0.0226 0.9266 LPLA (IV) -1.1552 0.5556 -0.0510 -1.4537 Male 0.3827 0.3967 0.0169 1.0676 Age 26-45 -2.0831 0.4015 -0.0920 -4.1399 Age 46-65 -1.9554 0.1813 -0.0864 -3.9873 Age 65< -2.3292 0.0443 -0.1029 -4.2654 Income <2.5K -1.1067 0.8787 -0.0489 -2.9189 Income 2.5-4.5 -0.5907 0.1009 -0.0261 -1.1898 Middle school -0.4051 0.5549 -0.0179 -1.0126 High school -0.6379 0.4008 -0.0282 -1.4779 Panel B
Wald test coefficients
Null Hypothesis Chi-squared statistic P-value
I = II 7.88 0.01 I = III 0.47 0.49 I = IV 6.23 0.01 II = III 4.25 0.04 II = IV 0.42 0.52 III = IV 4.01 0.05
27
TABLE 12: Robustness check perceived risk attitude.
Panel A shows the results of a robustness check using half of the dataset to check for consistency across the variables. Panel B shows the linear probability model (LPM) results
Panel A
Complete dataset (N = 1,115) Half dataset (N = 738)
Marginal effects z-Statistic Marginal effects z-Statistic
HPHA (I) 0.0351 1.7230 0.0105 0.5269 HPLA (II) -0.0276 -1.2155 -0.0149 -0.6937 LPHA (III) 0.0226 0.9266 -0.0218 -0.7026 LPLA (IV) -0.0510 -1.4537 -0.0331 -1.0723 Male 0.0169 1.0676 0.0000 -0.0018 Age 26-45 -0.0920 -4.1399 -0.0292 -2.1781 Age 46-65 -0.0864 -3.9873 -0.0458 -2.8901 Age 65< -0.1029 -4.2654 -0.0298 -1.7569 Income <2.5K -0.0489 -2.9189 -0.0069 -0.4988 Income 2.5-4.5 -0.0261 -1.1898 -0.0123 -0.8866 Middle school -0.0179 -1.0126 -0.0401 -2.9203 High school -0.0282 -1.4779 -0.0492 -1.7375 Panel B
Linear probability model
28
29
5. Conclusion
This study uses the logit model to investigate the role of financial literacy in determining the risk attitude of personal investors. The conclusions on the empirical research will answer the main question: What is the effect of financial literacy on the risk attitude of personal
investors?
To answer this question a main hypothesis is presented, supported by four different sub hypotheses:
H0: Financial literacy has no effect on risk attitude.
H1: Actual financial literacy has an effect on actual risk attitude. H2: Perceived financial literacy has an effect on actual risk attitude. H3: Actual financial literacy has an effect on perceived risk attitude. H4: Perceived financial literacy has an effect on perceived risk attitude.
H1: Actual financial literacy has an effect on actual risk attitude. The first hypothesis states that actual financial literacy has an effect on actual risk attitude. From the results in table 9, holding the perceived literacy the same, shows that the actual financial literacy has a strong effect on the actual risk attitude. When holding the perceived financial literacy at high and varying the actual financial literacy the marginal effects shift from positive to negative. The actual effect has a probability of 16.72% (8.2% + 8.52%). H1 is not rejected at the 5% significance level, since the variables have a significant effect on the risk attitude.
H2: Perceived financial literacy has an effect on actual risk attitude. Holding all other variables constant and varying only in perceived financial literacy shows that the effect is less strong than for the actual financial literacy is the case. The perception effect is, holding actual financial literacy at low, 4.28 % (12.82% - 8.52%). Since both variables, HPLA and LPLA, are significant, H2 is not rejected at the 5% significance level.
H3: Actual financial literacy has an effect on perceived risk attitude and H4: Perceived financial literacy has an effect on perceived risk attitude. The third and fourth hypotheses state that actual financial literacy has an effect perceived risk attitude. The test result in table 13 show that none of the four literacy variables has an significant effect on perceived risk attitude. H3 and H4 are accordingly rejected.
30
personal investors. For the perceived risk attitude the conclusion is different. Financial literacy, regardless the form of literacy, has no effect on perceived risk attitude. Concluded the answer for the main question of this study is that the effect of financial literacy depends on the form of literacy and risk attitude.
The results are in line with research as that financial literacy has an significant effect on risk attitude (Lusardi and Mitchell, 2007b, van Rooij, Lusardi and Alessie, 2007). The difference between actual risk attitude and perceived risk attitude can be allocated to the subject way of measuring perceived risk attitude. Respondents who completed the
questionnaire could be forgotten some risky choices they did in the past. Another cause of the deviation results could be the small sample size. Or the risk attitude is not normally
31
References
• Abreu, M., & Mendes, V. (2010). Financial literacy and portfolio diversification. Quantitative finance, 10(5), 515-528.
• Agnew, J. R., & Szykman, L. R. (2005). Asset allocation and information overload: The influence of information display, asset choice, and investor experience. The Journal of Behavioral Finance, 6(2), 57-70.Olsen, R. A. (1997). Investment risk: The experts' perspective. Financial Analysts Journal, 53(2), 62-66.
• Allgood, S., Walstad, W. (2012). The effects of perceived and actual financial literacy on financial behaviors. Available at SSRN 2191606.
• Attanasio, O. P., & Weber, G. (2010). Consumption and saving: models of intertemporal allocation and their implications for public policy (No. w15756). National Bureau of Economic Research.
• Barber, B. M., & Odean, T. (2000). Trading is hazardous to your wealth: The common stock investment performance of individual investors.Journal of Finance, 773-806. • Barber, B. M., & Odean, T. (2001). Boys will be boys: Gender, overconfidence, and
common stock investment.Quarterly journal of Economics, 261-292.
• Barber, B. M., & Odean, T. (2001). The internet and the investor.Journal of Economic Perspectives, 41-54.
• Behrman, J. R., Mitchell, O. S., Soo, C. K., & Bravo, D. (2012). How financial literacy affects household wealth accumulation.The American economic review,102(3), 300.
• Browning, M., & Lusardi, A. (1996). Household saving: Micro theories and micro facts. Journal of Economic literature, 1797-1855.
• Bruine de Bruin W, Parker AM, Fischhoff B. Individual differences in Adult Decision-Making Competence. Journal of Personality and Social Psychology. 2007; 92:938–956. [PubMed: 17484614]
• Bruine de Bruin, W., Vanderklaauw, W., Downs, J. S., Fischhoff, B., Topa, G., & Armantier, O. (2010). Expectations of inflation: The role of demographic variables, expectation formation, and financial literacy. Journal of Consumer Affairs, 44(2), 381-402.
• Cox, D. R., & Snell, E. J. (1989). Analysis of binary data (Vol. 32). CRC Press.
32
• Delavande, A., Rohwedder, S., & Willis, R. J. (2008). Preparation for retirement, financial literacy and cognitive resources. Michigan Retirement Research Center Research Paper, (2008-190).
• Diacon, S. (2004). Investment risk perceptions. Do consumers and advisers agree? The International Journal of Bank Marketing, 22, 180–198
• Diacon, S. (2004). Investment risk perceptions: Do consumers and advisers agree?. International Journal of Bank Marketing, 22(3), 180-199.
• Dominitz J, Hung AA, Yoong J. How Do Mutual Fund Fees Affect Investor Choices? Evidence from Survey Experiments. 2009 RAND Working Paper WR-653.
• Friedman, M., 1957. A Theory of the Consumption. Princeton, NJ: princeton university press.
• Gustman, A. L., Steinmeier, T. L., & Tabatabai, N. (2010). Financial knowledge and financial literacy at the household level (No. w16500). National Bureau of Economic Research.
• Hilgert, M. A., Hogarth, J. M., & Beverly, S. G. (2003). Household financial
management: The connection between knowledge and behavior. Fed. Res. Bull., 89, 309.
• Hogarth, J. M., & Hilgert, M. A. (2002). Financial knowledge, experience and learning preferences: Preliminary results from a new survey on financial literacy. Consumer Interest Annual, 48(1), 1-7.
• Hung, A., Parker, A. M., & Yoong, J. (2009). Defining and measuring financial literacy.
• Huston, S. J. (2010). Measuring financial literacy. Journal of Consumer Affairs,44(2), 296-316.
• Jappelli, T., & Padula, M. (2013). Investment in financial literacy and saving decisions. Journal of Banking & Finance, 37(8), 2779-2792.
• Kimball, M. Cognition and financial literacy. Willis, R., editor. Jackson Hole, WY: Presentation at the Conference on Understanding Economic Decision-Making; 2008 Sep.
• Langer, E. J. (1975). The illusion of control. Journal of personality and social psychology, 32(2), 311.
• Lusardi, A., & Mitchell, O.S. (2007b, October). Financial literacy and retirement planning: New evidence from the RAND American Life Panel. MRRC Working Paper No. 2007-157.
33
• Lusardi, A., Michaud, P. C., & Mitchell, O. S. (2013). Optimal financial knowledge and wealth inequality (No. w18669). National Bureau of Economic Research.
• Maddala, G. S. (1986). Limited-dependent and qualitative variables in econometrics (No. 3). Cambridge university press.
• Modigliani, F., & Brumberg, R. (1954). Utility analysis and the consumption function: An interpretation of cross-section data. Franco Modigliani, 1.
• Parker, A. M., Bruin, W. B., Yoong, J., & Willis, R. (2012). Inappropriate confidence and retirement planning: Four studies with a national sample.Journal of behavioral decision making, 25(4), 382-389.
• Remund, D. L. (2010). Financial literacy explicated: The case for a clearer definition in an increasingly complex economy. Journal of Consumer Affairs,44(2), 276-295.
• Sachse, K., Jungermann, H., & Belting, J. M. (2012). Investment risk–The perspective of individual investors. Journal of Economic Psychology, 33(3), 437-447.
• Van Rooij, M., Lusardi, A., & Alessie, R. (2011). Financial literacy and stock market participation. Journal of Financial Economics, 101(2), 449-472.
• Weber, E. U., & Milliman, R. A. (1997). Perceived risk attitudes: Relating risk perception to risky choice. Management Science, 43(2), 123-144.
34
Appendixes
Basic Literacy questions
Questions Answers
(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,-
(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,-
(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
(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
(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?
35
Advanced literacy questions
Questions Answers
(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
(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
(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
(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
(10) If the interest rate falls, what should happen to bond
prices?
(i) Rise (ii) Fall
(iii) Stay the same (iv) None of the above
36 (11) Buying a company
stock usually provides a safer return than a stock mutual fund. True or false?
(i) True (ii) False
(12) Stocks are normally riskier than bonds. True or
false? (i) True (ii) False (13) 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
(14) Normally, which asset displays the highest fluctuations over time?
(i) Increase (ii) Decrease (iii) Stay the same
(15) When an investor spreads his money among different assets, does the risk
of losing money:
(i) Increase (ii) Decrease (iii) Stay the same
(16) If you buy a 10-year bond, it means you cannot sell it after 5 years without incurring a major penalty.
