Printer Friendly

Consumer biases and competences in company stock holdings.

Consumers may make inefficient investment decisions in holding their own-company stocks because of several biases documented by the literature of behavioral economics, such as company stock bias, home bias, and mere exposure effect. Consumers with high levels of competences may be likely to overcome biases and make effective investment decisions. Using data from the 2004 and 2007 Surveys of Consumer Finances, evidence suggests the existence of the biases and competences, whereas a higher level of consumer competence can partially offset the influence of consumer biases.

**********

It is well documented that low diversification is an inefficient consumer investment decision. However, of approximately 23 million defined contribution (DC) retirement plan participants who have access to hold company stock, 5.3 million participants hold more than 60% of account balances in company stocks (Mitchell and Utkus 2004). More recently, Vanguard Group (2007) found that four in ten consumers continue to hold a concentrated position in their employer's stock in DC plans, even after seeing high profile examples such as the Enron and WorldCom cases that demonstrated individual company stock risk. (1) Laise (2009) reported that even during the subprime mortgage crisis and subsequent global financial crisis that started in 2008, "... at a time of extreme market uncertainty, many 401(k) participants are making a risky move: loading up on the stock of their employer."

Much research has documented that own-company stock investment will result in an underdiversification problem. For example, Meulbroek (2005) estimated the cost of holding own-company stock in 401(k) plans relative to a well-diversified equity portfolio of the same risk. She found that with an assumed investment horizon of 10 years and 25% of plan assets in company stock, consumers would sacrifice 42% in value when they invest in own-company stock that could otherwise have been earned in a diversified portfolio of the same risk. The problem of inefficient investment decisions may even be more serious when the consumers' current employment and career future depend on the same company. In the worst-case scenario, when the company fails, consumers can lose their income and much of their retirement wealth simultaneously.

Consumers tend to use heuristics in complex financial decision making (see Kozup and Hogarth 2008 for a review). In particular, consumers prefer investing in a familiar company and feel comfortable with investing in a business that is visible to them (Huberman 2001). Thus, on one hand, consumer biases may contribute to ineffective investments in terms of company stock holdings. As described by D'Antona (2008, p. 1), "some employees may feel the need for security and find that investing in their own company is a good thing. It is like putting money into a money market fund for them... They are so afraid of buying everything else; they know at their company they have a job and a desk, and see the S&P index or fund as much less tangible than their own company." Therefore, employer's company stock can be one type of familiar investment (Huberman 2001). Compared to other equities, it seems to be treated as a unique asset class with a quite different perceived risk distribution from the consumers' points of view (Benartzi 2001; Benartzi and Thaler 2001; Benartzi et al. 2007). In addition, familiarity bias is documented to be related to geographical approximation (e.g., home bias effect, Grinblatt and Keloharju 2001; Huberman 2001; Ivkovic and Weisbenner 2005) and can be induced by repeated exposure to stimuli (e.g., mere exposure effect, Zajonc 1968). On the other hand, consumer competences

documented by other studies (e.g., Graham, Harvey, and Huang 2009) may help consumers make effective decisions in terms of diversifying out of company stocks.

To date, the direct evidence of consumer biases (company stock bias, home bias, and mere exposure effect) and competence effects on consumer company stock holdings is missing in the literature and we hope to fill this research gap. Based on national, individual-level portfolio data from the 2004 and 2007 Surveys of Consumer Finances (SCFs), we examine own-company stock holdings in regular equity accounts, which, unlike DC plans, have few rediversification and trading constraints. Also, since the 2004, SCF has started to provide information regarding company stock holdings in both regular equity and pension accounts, we can investigate the interaction relationship in both accounts.

Due to possible biases, consumers may make systematic errors in assessing the risks of their company stocks, resulting in an inefficient investment decision. More importantly, a detailed understanding of behavioral factors and the mediating effects of consumer competences on consumer biases can help develop effective public policies, decreasing the probability that consumers turn to government or social resources as a last resort if their employers fail.

HYPOTHESIS DEVELOPMENT

Previous studies have focused on concentrated company stock holdings in DC plans (e.g., VanDerhei et al. 1999; Clark et al. 2000; Benartzi 2001; Mitchell and Utkus 2004) and have linked the concentrated holdings to various plan design features. However, unlike DC plans, company stock holdings in regular equity accounts are not subject to any rediversification and trading constraints.

Consumer Biases

Empirical evidence on consumer financial decision-making documents that consumers favor investments they are more familiar with. There are many implications of this tendency to invest in familiar securities in the context of own-company stock holdings. First, Huberman (2001) documented that employer's company stock can be one type of investment in the most familiar. Several studies have found that consumer perceived risk distribution of employer's company stock could be quite different from its true risk distribution (e.g., Benartzi 2001; Benartzi and Thaler 2001; Benartzi et al. 2007). This company stock bias could distort the relationship between consumer risk attitude and risk-taking behavior in the context of company stock holdings.

Second, consumers have shorter perceived geographical distance to own-company stock than to other equities and may be subject to the preference for investing in securities with geographical approximation, which by definition is called home bias effect (e.g., French and Poterba 1991; Coval and Moskowitz 1999). Many studies have provided explanations for the preference toward geographical-related investment based on consumers' fear of change and of the unfamiliar, or even directly examine familiarity bias by using home bias measures (e.g., Grinblatt and Keloharju 2001; Huberman 2001; Zhu 2003; Cao et al. 2007).

Third, many companies repetitively expose their employees to company stock via various programs, such as company stock in DC plans, resulting in mere exposure effect. Mere exposure effect is thought to increase the stimulus familiarity that, in turn, increases the stimulus liking (e.g., Berlyne 1970; Stang 1973). Therefore, there is a need to examine the effects of consumer biases on consumer company stock holdings via the indicators that are related to familiarity bias: consumers' risk-taking attitude, home bias, and mere exposure effect.

Company Stock Bias

Holding company stock in 401(k) plans is documented to be costly. For example, Poterba (2003) used a certainty equivalent approach to estimate the certainty equivalent value of a portfolio invested in company stock and found that even for moderately risk-averse consumers with substantial financial assets outside their 401(k) plans, holding only company stock in 401(k) plans would lead to 22% reduction in the certainty equivalent value, compared to the certainty equivalent of a portfolio invested in the S&P 500. Based on the Capital Asset Pricing Model, Meulbroek (2005) found that consumers would sacrifice an average 42% in value when they invested in company stock, as opposed to a diversified stock portfolio. As a result, company stock may not be an appropriate vehicle for investors who do not want to take more than the average financial market risk (e.g., S&P 500 index).

However, Benartzi and Thaler (2001) documented that company stock was not a substitute for other equities but a unique asset class. They compared behaviors of consumers with and without access to company stock in their 401(k) plans. They found that consumers who participated in the plans with access to company stock invested 42% in company stock and split the remaining 58% evenly between equities and bonds, whereas the consumers in the plans with no access to company stock had approximately 50-50 split between equities and bonds. They concluded that the result supported that company stock was treated as a special asset class different from other equities.

Furthermore, for this unique asset class of company stock, consumers seem to have quite different perceived risk. Based on a Sample including S&P 500 firms, Benartzi (2001) found that consumers' allocation to company stock was positively related to the past 10-year returns of company stock, indicating consumers tend to extrapolate past returns into future even though past returns do not predict future returns. Benartzi et al. (2007) also conducted surveys of more than 500 employees from roughly 100 different companies. They found that 42% of the respondents thought their company stock had the "same level of risk" as a diversified fund of the S&P 500. However, the average standard deviation of their actual holdings was 36% versus the market's 18%. In addition, 22% claimed that their company stock was "less risky." Yet, these stocks had an average standard deviation of 31% versus the market's 18%. Therefore, consumers systematically underestimate the risk of company stock. Because company stock bias leads to underestimation of the risk of own-company stock, consumers would demonstrate discrepancies between risk-taking attitudes and behaviors in terms of company stock holdings compared to their risk-taking attitudes and investment behaviors with respect to other equities. Thus, we form the following hypothesis.

Hypothesis 1-1 (Hl-1): Consumer risk-taking attitudes are not associated with corresponding risk-taking behaviors such as holding company stocks.

Home Bias

The equity home bias puzzle is the term given to describe the fact that consumers and institutional investors in most countries hold modest amounts of foreign equity. This is puzzling since observed returns on international portfolios suggest substantial benefits from international diversification. The home bias in equities was first documented by French and Poterba (1991) and has been identified as one of the six major puzzles in international macroeconomics (Obsfeld and Rogoff 2000). Coval and Moskowitz (1999) further showed that home bias was not limited to international portfolios, but that the preference for investing in securities with geographical approximation. Specifically, they showed that U.S. investment managers exhibited a strong preference for locally headquartered firms.

It is possible that home bias in equities can be attributed to capital immobility in an international environment. However, Van Nieuwerburgh and Veldkamp (2005) had some concerns regarding the capital immobility assumption. Other studies related home bias or geographic bias to familiarity bias. Huberman (2001) tested the familiarity bias due to geographic bias, and found that shareholders of a Regional Bell Operating Company tended to live in the area which it served. Grinblatt and Keloharju (2001) investigated the influence of geographic distance as well as language and culture on consumers using a unique data set of Finland. They documented that consumers were more likely to hold and trade the stocks of Finnish firms that were located close to them and that had the same language or cultural background. Zhu (2003) explored the geographic bias based on a large data set of individual accounts from a discount broker and found that the geographic bias stemmed from familiarity. Ivkovic and Weisbenner (2005), however, found that geographic bias could stem from consumers' ability to exploit local asymmetric information. Based on a data set of Swedish people, Massa and Simonov (2006) documented that people were more likely to have a concentrated holding of stocks that they were geographically or professionally close to or that they had held for a long time. They further showed that this tendency was information driven. Cao and colleagues (2007) referred to the preference toward geographically and culturally related choices as one of the "focal choice alternatives," and offered a unified explanation for these effects based upon consumers' fear of change and of the unfamiliar.

Consumers have shorter perceived geographical distance to owncompany stock than to other equities and may be subject to the preference for investing in securities with geographical approximation, which by definition is called home bias effect. We assume that consumers who prefer to invest in stocks in their home country by definition are subject to home bias, leading to a concentrated holding in company stock to which consumers are geographically closer than they are to other equities, which results in the following hypothesis.

Hypothesis 1-2 (Hl-2): Consumers who tend to adhere to domestic investment will hold a concentrated holding in own-company stock.

Mere Exposure Effect

The mere exposure effect is a psychological phenomenon first studied by Zajonc (1968) and later well known to advertisers: it is the formation of a positive affective reaction to repeated exposure to a stimulus, even in the absence of awareness. Thus, mere exposure effect is thought to increase the stimulus familiarity that, in turn, increases the stimulus liking (e.g., Berlyne 1970; Stang 1973). In application, consumers may. be influenced by the mere exposure effect when advertisers take advantage of consumers' fear of change and of the unfamiliar by repeatedly exposing consumers to the name of a brand. Advertisers hope that consumers will use their brand as a default choice when facing uncertainty about similar products. In the context of company stock holdings, many companies repetitively expose their employees to company stock, resulting in mere exposure effect; consumers react affectively to company stock by holding concentrated company stock holdings. Company stock offerings in 401(k) plans can be one type of repeated exposure.

Specifically, in 401 (k) plans, employer matching contributions can be either directed to company stock at the employers' discretion, invested at the participants' discretion, or a combination of both. In addition, some employers offer company stock as one of the investment options in their 401(k) plans and consumers can choose whether or not to invest their discretionary contributions in company stock. In addition, several plan designs including employer-imposed restrictions and company stock match may have endorsement effect (Liang and Weisbenner 2002). As a result, consumers who hold company stocks in their 401(k) plans may subject to mere exposure effect. We then form the following hypothesis.

Hypothesis 1-3 (Hl-3): Consumers who own company stock in their 401(k) plans tend to have a higher level of company stock holdings in regular equity accounts.

Consumer Competences

Heath and Tversky (1991) provided experimental research to illustrate that consumers' preference for their own judgment bet over the matchedchance lottery was affected by their competence, where competence is defined as the subjective knowledge level in a certain area. They argued that competence can be enhanced by general knowledge and experience. Similarly, Perry and Morris (2005) found that consumers' level of perceived control over outcomes would affect their propensity to save, budget, and control spending.

For decision-making in the financial setting, it is analogously presumed that consumer competence is positively correlated with consumers' general knowledge and experience, including financial literacy and investment experience, where financial literacy is related to the ability to identify, locate, understand, evaluate, and use information effectively (Caissey 1990; Lee and Cho 2005). Using the UBS/Gallup Investor Survey, Graham, Harvey, and Huang (2009) found that consumers who felt competent would trade more often and have a more internationally diversified portfolio. The competence measure in their analysis was modeled as a function of gender, education, income, and the size of total investment. Graham, Harvey, and Huang found that consumer competence increased with males, level of education, income, and the size of total investment. Because consumer competence variables are not available from SCFs, we used three proxy variables, including education, income, and share of other stocks in financial assets. We assume that higher education, higher income, and larger shares of other stocks in financial assets are associated with higher financial competence and form the following hypotheses.

Hypothesis 2-1 (H2-1): Consumers with a higher education level would have a more diversified stock portfolio and thus a less concentrated holding in company stock.

Hypothesis 2-2(H2-2): Consumers with higher income would have a more diversified stock portfolio and thus a less concentrated holding in company stock.

Hypothesis 2-3(H2-3): Consumers who hold a higher level of publicly traded stocks in their financial assets (not including own-company stock) would have a more diversified stock portfolio and thus a less concentrated holding in company stock.

Financial literacy can be acquired not only through formal education, but informal information intermediaries, such as the Intemet, friends, and financial advisors. Rappa (2003) found that consumers increasingly relied on information intermediaries for their decision making. Coleman, Warren, and Huston (1995) and Lee and Cho (2005) found that consumers preferred human sources for information acquisition/integration to nonhuman sources for decisions involving high degrees of uncertainty and importance. More specifically, enormous and complicated financial information can create information overload for consumers, some of whom simply walk away from making a decision (Gourville and Soman 2005). In addition, the large variety of financial products even makes it a hard task for regulators or educators to construct a standardized decision aid that is easily understandable for consumers (Kozup and Hogarth 2008). Even when consumers have the ability to comprehend the financial information, they might not want to get involved in the details of financial decision-making and would rather count on someone else to do that, thus remaining financially aliterate (Rotfeld 2008). As a result, financial services advisors or counselors can help to explain financial information and provide suggestions.

For example, Hershey and Walsh (2000) found that experts were likely to engage in a consistent level of goal-oriented, efficient financial planning. Novice investors, however, likely focus on the salient opinions or information (e.g., brand name). Thus, consumers who raise their level of financial literacy by obtaining advice from professionals become more competent and more likely to make efficient investment decisions, leading to a less concentrated holding in company stock, which results in the following hypothesis.

Hypothesis 2-4(H2-4): Consumers who seek advice from professionals would have a more diversified stock portfolio and thus a less concentrated holding in company stock.

METHOD

Data

Data obtained from the 2004 and 2007 Surveys of Consumer Finances (SCFs), sponsored by the Federal Reserve Board. Bucks, Kennickell, and Moore (2006) and Bucks and others (2009) provide detailed descriptions of the changes in U.S. family finances based on the 2004 and 2007 SCFs, respectively. The SCF has been regularly conducted every three years since 1983, but the 2004 and 2007 SCFs are the first ones to provide information on company stock holdings in both regular equity and pension accounts. In particular, in the 2004 and 2007 SCFs, respondents were asked the following question: "What is the total market value of your stock in the company (where you work) [in regular equity accounts]?" (X3922). In addition, respondents were asked the following information regarding company stock holdings in account-type plans: "About what percent of[the balance of your pension account] is in stocks?" (X11038) "About what percent of this stock is in company stock?" (X11039).

Equities can be held in pension accounts and nonpension accounts. In the SCFs, equity holdings in nonpension accounts include direct stock holdings, indirect holdings through stock mutual funds, stocks or stock mutual funds in IRA/Keoghs accounts, trusts, and managed accounts. Equity holdings in nonpension accounts are referred to as regular equity accounts and were the focus herein. Company stock holdings in pension plans are subject to many plan restrictions and the information is unavailable in the 2004 and 2007 SCFs, while company stock holdings in regular equity accounts are free of any restrictions. We first restricted our sample to respondents in the 2004 and 2007 SCFs who were working at the time of the studies (X6670 = 1). Furthermore, we selected respondents who held a positive percentage of company stock in their regular equity accounts. After applying these selection criteria, the final sample included 341 respondents in 2004 and 312 respondents in 2007. The population defined by the weights provided in the SCFs is 6,054,056 households for the 2004 sample and 6,862,912 households for the 2007 sample. In addition, the imputation technique was implemented to deal with the missing SCF survey data. Therefore, the reported coefficient estimates in the analysis of variance (ANOVA) and multivariate regression analyses were obtained by running the Repeated-Imputation Inference procedure (RII) suggested by Montalto and Sung (1996). In addition, following Lindamood, Hanna, and Bi's (2007) suggestion, we used a weighted sample for descriptive analyses and an unweighted sample for ANOVA and multivariate analyses.

Variables

Dependent Variables

The decision-making process for a consumer to invest in equity-related assets may be different from the decision-making process to invest in company stock. For example, consumer's risk aversion would affect portfolio decision making. A risk-averse consumer may have a small share of his or her financial assets held in equity-related assets, but hold all the equity-related assets in company stock due to behavioral biases. A small share of financial assets held in equity-related assets can be consistent with one's risk aversion (e.g., Friend and Blume 1975), while a large share of equity-related assets in company stock for a risk-averse consumer cannot be explained by risk aversion, but by other factors we are going to explore. In order to separate the decision making of investing in equity-related assets from that of investing in company stock, we defined the dependent variable as the company stock value in nonpension accounts divided by equity-related financial asset value in nonpension accounts. In this measure, the dependent variable was a proxy of diversification. A lower proportion of company stock was associated with diversifying out of company stock and thus a better investing behavior in terms of own-company stock holdings.

To be more specific, in the 2004 and 2007 SCFs, the numerator of the dependent variable was the market value of company stock in regular equity accounts (X3922). For the denominator, equity-related financial assets in regular equity accounts included direct stock holdings, indirect holdings through stock mutual funds, IRA/Keoghs accounts, trusts, and managed accounts. Specifically, direct stock holdings included company stock and other publicly traded stocks. Stock mutual funds included the full value of stock mutual funds and 1/2 of the value of combination mutual funds. Stocks invested in IRAs/Keogh accounts included the full value of the accounts if mostly invested in stocks, 1/2 of the value if split between stocks/bonds or stocks/money markets, and 1/3 of the value if split between stocks/bonds/money markets. Other managed assets with equity interests, including annuities and trusts, included the full value of the accounts if mostly invested in stocks, 1/2 of the value if split between stocks/mutual funds and bonds/CDs, or mixed/diversified, and 1/3 of the value if "other." If the dependent variable of company stock as a percentage of equity-related financial assets is high, it means that a consumer holds a concentrated company stock holding in the regular equity accounts, If the ratio is low, it means that a consumer diversifies out of company stock.

One may be concerned that owning employer's company stock is not always an employee's choice as, for example, with an employee who is rewarded for performance through a stock bonus plan. However, consumers who acquire company stock not by choice can sell some holdings for rediversification, given the high cost of holding own-company stock relative to a well-diversified equity portfolio of the same risk. Consumers who fail to rediversify company stock holdings in regular equity accounts do not trade away from their current ownership positions, which is referred to status quo bias (e.g., Samuelson and Zeckhauser 1988; Fox and Tversky 1995). Ritov and Baron (1992) claimed that familiarity or fear of the unknown is a factor that causes the status quo bias and Cao et al. (2007) documented that consumers' favor of status quo choices is under the broad category that consumers prefer the familiar and fear change and the unknown. As a result, our familiarity bias-related variables make sense to provide underlying rationale for concentrated company stock holdings (i.e., underdiversification) in regular equity accounts.

Independent Variables

Specifications of variables in the SCFs used to measure the independent variables are shown in the Appendix. The contents are described below.

Risk Tolerance

Four dummy variables were created to reflect the risk-taking attitude of a consumer: (1) Substantial risk, willing to take substantial financial risks expecting to earn substantial returns; (2) Above average risk, willing to take above average financial risks expecting to earn above average returns; (3) Average risk, willing to take average financial risks expecting to earn average returns; and (4) No risk, not willing to take any financial risks.

Home Bias

The tendency of consumers to invest in domestic investments is referred to home bias and was measured as follows: "Of your stock, is any of its stock in a company headquartered outside of the United States?" We set the dummy variable, Home bias, equal to one if a consumer did not hold any stock in a company headquartered outside of the United States, and zero otherwise.

Mere Exposure Effect

One variable was used to measure mere exposure effect. Company stock in DC plans was the company stock as a share of total equities in 401 (k) plans.

Competence Variables

The competence hypothesis was measured by four variables. One was the education level of a consumer. A dummy variable, College, was set to one if a consumer had some college education, and zero otherwise. Also, a dummy variable, Graduate, was set to one if a consumer had some graduate study, and zero otherwise. The omitted category was Lower than college. The second variable was related to consumer income level. A dummy, Head's total earnings (20075)from job in top 25%, was set to one if the income of a consumer was ranked in the top quartile within one's industry, and zero otherwise. It was presumed that a consumer who had a relatively high income compared to his/her counterparts was more financially competent. In addition, in the SCFs, respondents were asked "Do you own any stock which is publicly traded? What is the total market value of this stock?" Therefore, the third variable, Other Stock/financial assets, was defined as the market value of publicly traded stocks minus market value of company stock, and then the difference was divided by total financial assets. A consumer who held a higher share of publicly traded stocks in financial assets was assumed to have more stock investment experience. Fourth, in regard to acquiring financial literacy via informal information intermediaries, we included a dummy, Seek professional advice, set to one if a consumer made investment decisions by obtaining advice from professionals, including bankers, brokers, financial planners, dealers, and insurance agents, and zero otherwise.

Several control variables were included in the analyses. We controlled for whether consumers had stock options or were offered Employee Stock Ownership Plans or Employee Stock Purchase Plans. Plans was set to one if a respondent had financial options granted by his/her employer or if the employer offered an Employee Stock Ownership Plan or Employee Stock Purchase Plan, and zero otherwise. We also controlled for inheritance/gift opportunity. Have received an inheritance or gift dummy variable was set to one if a respondent has ever received an inheritance, or been given substantial assets in a trust or in some other form, and zero otherwise. Two control variables are relevant to work. A dummy variable, Large firm size, was set to one if the number of employees in the company where a consumer worked was more than 499, and zero otherwise. A continuous variable, Work year, was measured by a consumer's number of years of work with the current company.

Regarding the race/ethnicity variable, if a consumer chose "Black/ African American" or "Hispanic/Latino" when presented with a list of racial/ethnic categories, Respondent Black or Hispanic was a dummy variable set equal to one, and if a consumer chose the remaining "White" or "Other" when presented with a list of racial/ethnic categories, this race dummy is set to zero. In SCFs, Asian is classified as "other." The rationale to use this collapsed category is that according to U.S. Census Bureau (2008), 53 percent of Asians in the United States had a bachelor's degree or more education. For non-Hispanic whites, it was 33 percent; for blacks; it was 20 percent; and for Hispanics, it was 13 percent. Thus "Other" group is more close to the "White" group in terms of higher education. In addition, financial attitudes and sophistication might be related to socialization and cultural background, so being a member of a racial/ethnic group with low income and education might affect choices beyond what would be expected on the basis of the individual respondent's education.

The SCFs collect attitude variables for the respondents, but reverse all the variables in the data set that originally referred to "respondent" and "spouse/partner" and all codes that contain the same references based on the rule that the male is always designated as the head for mixed-sex couple households and the older individual is designated as the head in same-sex couple households. Consistent with the argument of Lindamood, Hanna, and Bi (2007), we used the respondent's sex, instead of the head's sex, to define whether a consumer was male, where Male respondent was a dummy equal to one if a respondent was male, and zero otherwise. Since Graham, Harvey, and Huang (2009) found that males had higher consumer competence than females, we then controlled the Male respondent dummy in the model. In addition, couple households may have financial resources different from noncouple households, leading to different investment behavior. Thus, we included a dummy Couple household, equal to one if the household type was couple household and zero for noncouple households.

Mediating Effects of Consumer Competences on the Influence of Consumer Biases

Previous studies have used hierarchical regression to examine the mediating effects based on sets of independent variables (e.g., Grable, Park, and Joo 2009). We also used the same approach to examine whether higher levels of consumer competences can offset the influence of consumer biases. The order with which we entered the variables was as follows: control variables (Model 1), consumer bias variables (Model 2), and consumer competence variables (Model 3). A series of regression equations were estimated. The increment in [R.sup.2] due to the incremental predictors from one model to the next model was examined by the F-test and was described as follows:

F = ([R.sup.2.sub.F] - [R.sup.2.sub.FR)/(f - r)/(1-[R.sup.2.sub.F])/(n-f -1),

where [R.sup.2.sub.F] = variance explained by full model,

[R.sup.2.sub.R] = variance explained by reduced model,

f = number of predictors in the full model,

r = number of predictors in the reduced model, and

n = observations.

RESULTS

Descriptive Statistics

Descriptive statistics of the sample used are presented in Table 1. SAS macros and Replicate Weight Files provided in the SCFs were used to compute standard errors as shown in Tables 1 and 2.

The dependent variable was the share of company stock in equity-related assets, conditional on positive company stock holdings. The mean value of the share of company stock in equity-related assets was 51.86%. We further looked into the cross table on percentages of company stock holdings using the weighted sample. The third column of Table 1 reports the results. Approximately 23% of respondents held no more than 10% of equity-related assets in company stock, and approximately 33% of respondents held more than 90% of their equity-related assets in company stock. The results indicate that most of the respondents are either in the low or high end. Similar results were obtained using the unweighted sample for a robustness check.

In addition, approximately 12% of respondents who held company stock in regular equity accounts were willing to take "no risk," and approximately 49% were willing to take "average risk." Because company stock is not an appropriate vehicle for investors who do not want to take more than average market risk, the results indicate that 61% (12% plus 49%) of respondents have a mismatch between risk-taking attitudes and company stock holdings in regular equity accounts. This result is broadly consistent with the 401 (k) research findings by Benartzi et al. (2007), who found that 64% of the 401(k) respondents to their Web site questionnaires misperceived that their company stock had the "same level of risk" or "less risky" than a diversified fund of the S&P 500, while these company stocks had risk far above that of the market index.

Regarding the behavioral characteristics of the respondents, approximately 92% of the respondents did not hold any foreign stocks. In addition, the average percentage share of company stock in 401(k) plans was 7 percent.

Regarding competence variables, the highest level of education of the respondent was some college but no graduate education for 26% of the respondents, and 57% of the respondents had some graduate education. Approximately 59% of the respondents were in the top-quartile income category within their industry. In addition, the average percentage of other publicly traded (domestic) stocks in financial assets was 9%. Approximately 32% of the respondents obtained financial advice from professionals during their investment decisionmaking.

Regarding controlled variables, approximately 33% of the respondents were offered financial options, Employee Stock Ownership Plan or Employee Stock Purchase Plan, and about 26% of the respondents had ever received an inheritance or been given substantial assets in a trust or in some other form. Sixty-six percent of the respondents worked for large firms with at least 500 employees, and the average number of working years with the current company was approximately 10. Fifty-six percent of the respondents in our sample were male, 10% were Black/Hispanic group, and 74% belonged to couple households.

Results of a preliminary check on whether consumers had different risk tolerance attitudes toward company stock and equities provide empirical support for company stock bias; consumers view company stock as a unique asset class different from equity, not a substitute for equity (Benartzi and Thaler 2001, 2007).

Panel A of Table 2 indicates that consumers who did not want to take any financial risk at all held the lowest level of equities as a percentage of financial assets (10.45%), but held the highest level of company stock as a percentage of financial assets (39.75%). This descriptive result was supported in the formal regression analysis in the Panel B of Table 2. In this regression, we collapsed Above average risk and Substantial risk into a single dummy and this collapsed dummy was the omitted category. Thus, compared to consumers who were willing to take substantial financial risk and above average risk, those who were willing to take No risk held a statistically significantly lower percentage of equity-related assets in financial assets, but held a statistically significantly higher percentage of company stock in financial assets. In addition, compared to consumers who were willing to take substantial financial risk and above average risk, those who were willing to take Average risk held a statistically significantly lower percentage of equity-related assets in financial assets, but held a statistically significantly higher percentage of company stock in financial assets. Both magnitudes were smaller than those of No risk consumers.

Therefore, extremely risk-averse consumers invested a small percentage of equity-related assets in their financial assets, but held a high percentage of company stock in their financial assets, as did those who were willing to take only average financial risk. This finding supports the argument that consumers treat company stock as a unique asset class different from equity, not a substitute for equity, and underestimate the risk of own-company stocks.

Hypothesis Tests

To assess the relationship between behavioral factors and consumer company stock holdings in regular equity accounts, we used analyses of variances (ANOVAs) and multivariate analyses. For each of the ANOVAs, consumer bias and competence variables were used as classification variables. Own-company stock as a percentage of equity-related assets was the focal variable. F-statistics in ANOVAs tested the overall model to determine whether there is a difference in means between different levels of each of the classification variables. As the rejection of the null hypothesis does not specifically tell us which means are different, significant differences between pairs of different levels on each of the classification variables were examined using least significant difference (LSD) tests. The ANOVAs were unweighted and revealed several significant relationships between consumer bias/competence and company stock holdings in regular equity accounts and are shown in Table 3. Findings of multivariate analysis are shown in Table 4.

Risk Tolerance

H1-1 suggests that there would be an inconsistency between risk attitudes and company stock holdings. In ANOVA, the LSD tests revealed that the percentage of company stock holdings was significantly higher when consumers did not want to take any financial risk at all compared to those who were willing to take average financial risk. In addition, consumers who were willing to take average financial risk had a significantly higher percentage of company stock holdings than those who were willing to take above average financial risk. Multivariate analysis found that consumers who were not willing to take any financial risk at all exhibited the highest percentage of company stock holdings compared to the omitted collapsed category of Substantial risk and Above average risk. The result is statistically significant, indicating an inconsistency between risk attitude and investment behavior in terms of company stock holdings for highly risk-averse consumers. Thus, H1-1 is supported.

Home Bias

In ANOVA in Table 3, the LSD tests revealed that when consumers held only domestic stocks (a proxy to home bias), the percentage of company stock holdings was significantly higher than when consumers held foreign stocks as well. In a multivariate analysis in Table 4, consumers who held only domestic stocks were found to have a higher percentage of company stock holdings than those who held foreign stocks as well and the effect is significant. Both the ANOVA and multivariate results support H1-2 that consumers who do not invest in stocks outside home country hold a higher level of company stock holdings in regular equity accounts.

Mere Exposure Effect

With regard to the mere exposure effect, H1-3 focuses on company stock holdings in 401 (k) plans and suggests that consumers who own high levels of company stock in their 401(k) plans due to various plan designs would hold a concentrated holding in company stock in regular equity accounts as well. Our findings in ANOVA LSD tests revealed a significant difference in company stock holdings in regular equity accounts when consumers' holdings in company stock in DC plans were ranked in the top third, as compared to those who were ranked in the bottom third or in the middle third. In the multivariate analysis in Table 4, consumers who held a higher level of company stock holdings in DC plans also held a statistically significantly higher level of company stock holdings in regular equity accounts. Therefore, H1-3 is supported.

Competence

H2-1, H2-2, H2-3, and H2-4 propose competence effects on company stock holdings. We hypothesized that consumers who are more financially competent would be more sophisticated in investing, resulting in holding a more diversified stock portfolio than a concentrated "home" portfolio to which own-company stock belongs. Our findings are as follows.

First, our ANOVA LSD tests found that consumers reported significant differences in company stock holdings across education levels. At-least-college-educated consumers held a less concentrated company stock holding than did lower-than-college-educated consumers. In addition, graduate-educated consumers held a less concentrated holding in company stock than did at-least-college-educated consumers. In the multivariate analysis in Table 4, the effect of Graduate was significantly negative, indicating that consumers who were graduate-educated held a less concentrated holding in company stock compared to consumers with lower-than-college education background (omitted category). H2-1 is supported for graduate-educated consumers.

Second, the LSD tests in the ANOVA showed that consumers who earned a relatively high income held a significantly less concentrated company stock holding than those whose incomes were not ranked in the top quartile within their industries. The multivariate analysis in Table 4 also revealed that consumers with a relatively high income compared to their counterparts in the same industries would hold a significantly less concentrated holding in company stock. H2-2 is supported.

Third, consumers who held a middle level (middle third) of publicly traded stocks in their financial assets reported a significantly lower level of company stock holdings than those who held a low level (bottom third). Similarly, consumers who held a high level (top third) of publicly traded stocks in their financial assets revealed a significantly lower level of company stock holdings than those who held a middle level (middle third). In the multivariate analysis in Table 4, consumers who held a higher level of publicly traded stocks in their financial assets held a significantly less concentrated holding in company stock. The result supports H2-3 that consumers who have experience in stock investing would hold a more diversified stock portfolio.

Fourth, consumers who obtained advice from professionals held a significantly less concentrated holding in company stock than those who obtained financial advice from other sources. The corresponding coefficient in multivariate analysis in Table 4 supported that consumers who obtained advice from professionals would hold a significantly less concentrated holding in company stock (H2-4).

Can Consumer Competences Offset the Influence of Consumer Biases?

Hierarchical regression examined whether a higher level of consumer competences can offset the influence of consumer biases. [R.sup.2.sub.F] (or [R.sup.2.sub.R]) was the average of five implicates in the RII procedure. The results are reported in Table 5.

When consumer bias variables as a block were added to the model (i.e., from Model 1 to Model 2), the incremental R-squared was 9.70%, which was significant ([F.sub.(4,641)] = 18.8260, p = .0000). In addition, when consumer competence variables as a block were further added to the model (i.e., from Model 2 to Model 3), the incremental R-squared was 15.20%, which was also significant ([F.sub.(5,636)] = 28.7234, p = .0000). Regarding consumer bias variables, the effect size of No risk on company stock holdings became smaller and the positive relationship between Average risk and company stock holding became insignificant when the competence variables as a block were added to the full model. The magnitudes of the Home bias effect also became smaller when the competence variables were added to the model. These results support the argument that a higher level of consumer competence appears to partially offset the influence of consumer biases. In particular, company stock bias for extremely risk-averse consumers becomes less severe when consumer competence is considered. Also, home bias effect will influence consumer company stock holdings, but this effect is lessened when consumer competence variables are added to the full model.

CONCLUSIONS AND IMPLICATIONS

Extremely risk-averse consumers demonstrate an inconsistency between risk-taking attitudes and behaviors in terms of company stock holdings. The result provides empirical support for the company stock bias proposed by Benartzi and Thaler (2001, 2007), which states that consumers tend to view company stock as a unique asset class different from equities and underestimate the risk of company stock. In addition, consumers who prefer the stocks of their home country also hold a high level of company stock to which they are geographically closer than to other equities. This finding can be placed in the context of home bias effect (e.g., Grinblatt and Keloharju 2001; Huberman 2001; Massa and Simonov 2006), direct empirical support that home bias is one of the contributing factors to consumers' concentrated company stock holdings. Finally, consumers who have company stock exposure in DC plans via a company stock matching policy or company stock investment option tend to have a concentrated holding in company stock in regular equity accounts as well. The result supports mere exposure effect (e.g., Zajonc 1968; Berlyne 1970; Stang 1973) and brings concerns that consumers are exposed to the risk of company failure in both regular and pension accounts.

However, consumers with high levels of competences may overcome biases and make effective investment decisions. Consistent with the findings of Lee and Cho (2005), Perry and Morris (2005) and Graham, Harvey, and Huang (2009), we assume that consumer competences in financial decision-making can be enhanced by general knowledge, relative income level, investment experience, and the assistance of professional advisors. The results suggest that consumers who are graduate-educated, who have a relatively high income compared to their counterparts in the same industries, or who have a higher percentage of publicly traded stocks held in financial assets tend to hold a less concentrated holding in company stock. In addition, consumers who seek advice from professionals in their investment decisions tend to hold a less concentrated holding in company stock compared to those who obtain advice from other sources. These results support the argument of Kozup and Hogarth (2008) that the third-party advice can be a crucial component for consumers in making optimal financial decisions, indicating that consumers who have competences tend to diversify out of own-company stock. Furthermore, consumer competence appears to partially offset the influence of consumer biases. In particular, the company stock bias for extremely risk-averse consumers becomes less severe when consumer competence is considered. The home bias tendency is lessened when consumer competence is added to the model.

The implication is that consumers who are extremely risk-averse are vulnerable to underdiversification problem in terms of own-company stock holdings. These consumers can be subject to external locus of control (Perry and Morris 2005), perceiving financial events as being under the control of luck and other people. They can also be subject to information overloading and simply decide not to make any decision. In addition, they can be subject to financial aliteracy (Rotfeld 2008--even if they can comprehend financial information, they seldom make decisions about money regardless of the potential costs if they make a mistake. These consumers tend to adopt salient opinions or information for financial decision-making, such as own-company name. As a result, they tend to hold a concentrated company stock holding in their regular equity accounts even though they do not want to take any financial risk at all.

The positive relationship between consumer home bias and concentrated company stock investment also offers a challenge to financial service providers and financial educators. The tendency of consumers to invest in familiar securities with geographical approximation will keep them underestimating the risk of own-company stock. Matters become more complicated as many companies repetitively expose their employees to company stock via various programs, such as company stock in DC plans, resulting in mere exposure effect, that is, consumer form a positive affective reaction to company stock. Our findings provide evidence of mere exposure effect on consumer behavior in company stock holdings. As a result, underestimating the risk of company stock can be a serious problem if consumers hold company stock not only in their regular equity accounts but also in pension accounts, resulting in exposure to excessive risk of company failure. Pension advisors and policy makers should be aware of this implicit communication/advising effect of repetitive company stock exposure on inefficient investment decision making. Designs and deliveries of DC retirement plans should be improved to alleviate the mere exposure effect to better protect consumers' interests.

This suggests that consumers with high levels of competence tend to hold a more diversified stock portfolio instead of a concentrated "home" portfolio to which own-company stock belongs. Furthermore, consumer competence in financial decision making appears to mediate the effect of consumer biases on company stock holdings, wherein some proxies for consumer competence are at least some graduate education, a relatively high income level, some publicly traded stock investing experience, and seeking assistance of professional advisors. Professional advisors can design adequate investment guidelines for consumer education and information to empower consumers to make effective investment decisions when they have access to own-company stocks. Special policies need to be developed to better protect vulnerable consumers who have low education and low income in company stock investing decisions.

The current trend of socioeconomic contexts emphasizes more individual responsibilities for consumer long-term economic security. Consumer policy makers and educators should beware of consumer vulnerabilities in long-term investing revealed in this and other behavioral economic studies and take appropriate measures to help consumers make effective investment decisions and ensure their long-term economic security.
APPENDIX 1
Description of Independent Variables

Description                       Questions in the SCFs

(1) Risk             Which of the statements on this page comes
  tolerance            closest to the amount of financial risk that
                       you are willing to take when you save or make
                       investments? (1) take substantial financial
                       risks expecting to earn substantial returns;
                       (2) take above average financial risks
                       expecting to earn above average returns; (3)
                       take average financial risks expecting to earn
                       average returns; (4) not willing to take any
                       financial risk

(2) Consumer bias
  variables          Of your stock, is any of it stock in a company
  Home bias            headquartered outside of the United States?
  Mere exposure      Account-type pension plans:
    effect
                     (1) About what percent of it is in stocks?

                     (2) About what percent of this stock is in
                     company stock?
                     (#la referred to first current job pension,
                       #lb referred to second current job pension,
                       and #I c referred to third current job
                       pension.)
                     If consumers whose pension plans were 401(k)
                       plans and who had positive balance in their
                       401(k) plans, the maximum value of
                       X11039(#la), X11139(#Ib), and X11239(#1c) was
                       recorded as the value of Company Stock in DC
                       plans, and zero otherwise.

(3) Competence
  variables          What is the highest grade of school or year of
  Education          college you completed?

  Income             (1) About how much do you earn before taxes on
                       your main job? (Also code frequency.)

                     (2) In addition to regular salary, how much do
                       you personally receive from the business
                       before taxes? (Also code frequency.)
                     Regular income and business income were added
                       together to estimate the income.
                     (3) What kind of business or industry do you
                       work in-that is, what do they make or do at
                       the place where you work(s)?

  Publicly traded    What is the total market value of this stock
    stocks           (publicly traded stocks)?
  Advice             What sources of information do you use to make
                       decisions about saving and investments?
                     How do you make decisions about saving and
                       investments?
                     (Code all that apply: code responses in the
                       order they are given.)
                     The first response was considered and 10 =
                       banker; 11 = broker; 12 = financial planner;
                       23 = store dealer; 24 = insurance agent

(4) Control
  variables          Some employers give their employees financial
  Stock option/        options that can be used to purchase company
    stock              stock at a later time. During the past year,
    purchase           has your current employer given you any of
    plans              these, either as a regular part of your
                       compensation, or as a bonus?
                     Are you included in any pension, retirement, or
                       tax-deferred savings plans connected with the
                       job you just told me about? What kinds of
                       plans are these?
                     X6712 = 7 Stock purchase/ESOP (Employee
                       Stock Option Plan)
  Inheritance/       Including any gifts or inheritances you may
    gift               have already told me about, have you (or your
                       [husband/wife/partner]) ever received an
                       inheritance, or been given substantial assets
                       in a trust or in some other form? IF YES:
                       Please do not include inheritances from a
                       deceased spouse.
                     1. *YES 5. *NO

(5) Working
  variables          About how many employees work for this company
  Firm size            or organization, including all locations?
                       (PROBE: is it fewer than 10, 10-19, 20-99,
                       100-499, or 500 or more?)
  Working year       How many years have you worked in this
                       business? (Code number of years.)

(6) Demographic
  variables          Which of these categories do you feel best
    Race/              describe you: (white, black or African
    ethnicity          American, Hispanic or Latino, Asian, American
                       Indian or Alaska Native, Hawaiian Native or
                       other Pacific Islander, or another race?) For
                       the public data set, for x6809, Asian/American
                       Indian /Alaska Native/ Native Hawaiian /
                       Pacific Islander were combined with Other.

  Sex                If the "respondent" and "spouse/partner" were
                       not reversed (x8000 = 5), the respondent's sex
                       (X8021) is used. On the other hand, if the
                       "respondent" and "spouse/partner" were
                       reversed (x8000 = 1), the spouse/partner's sex
                       (X103) is used to determine the Male dummy.

  Household          Are you currently married or living with a
    type               partner, separated, divorced, widowed, or have
                       you never been married?

Description                Variable #               Variable Name
                           in the SCFs

(1) Risk             X3014                     Four dummies:
  tolerance                                    Substantial risk
                                               Above average risk
                                               Average risk
                                               No risk

(2) Consumer bias
  variables          X7640                     Home bias dummy
  Home bias
  Mere exposure                                Company stock in DC
    effect                                     plans
                     (1) X11037 (#la)/
                       X 11137(# 16)/
                       X 11237(# 1 c))
                     (2) X11039(#la)/
                       X11139(#lb)/
                       X11239(#1c)

(3) Competence
  variables          Less than college         College = yes and
  Education            X5901 < 13              Graduate = yes dummies
                     College 12 < X5901 < 17
                     Graduate X5901 > 16
  Income             (1) X4112 (X4113)         Head's total earnings
                                               (2007$) from job is in
                                               top 25%
                     (2) X4131 (X4132)

                     (3) X7402 (This varia-
                     ble was used for total
                     earnings ranking within
                     one's industry)
  Publicly traded    X3915                     Other stock/financial
    stocks                                     assets
  Advice             X7112 = 10,11,12,23,24    Seek professional
                                               advice = yes

(4) Control
  variables          X6797                     Plan (employer offers
  Stock option/                                stock option plan) =
    stock                                      yes
    purchase
    plans

                     X6712 = 7

  Inheritance/       X5801                     Have received an inhe-
    gift                                       ritance or gift = yes

(5) Working
  variables          X4114                     Large firm size = yes
  Firm size                                    if 499

  Working year       X4115                     Work year

(6) Demographic
  variables          X6809                     Respondent Black or
    Race/                                      Hispanic = yes
    ethnicity

  Sex                X8000 and X8021 (X103)    Male respondent = yes

  Household          X8023 (X105)              Couple household = yes
    type
                     X7020


REFERENCES

Benartzi, Shlomo. 2001. Excessive Extrapolation and the Allocation of 401(k) Accounts to Company Stock. Journal of Finance, 56 (October): 1747-1764.

Benartzi, Shlomo and Richard H. Thaler. 2001. Naive Diversification Strategies in Retirement Saving Plans. American Economic Review, 91 (March): 79-98.

Benartzi, Shlomo and Richard H. Thaler. 2007. Heuristics and Biases in Retirement Savings Behavior. Journal of Economic Perspectives, 21 (Summer): 81-104.

Benartzi, Shlomo, Richard H. Thaler, Stephen P. Utkus, and Cass R. Sunstein. 2007. The Law and Economics of Company Stock in 401(k) Plans. Journal of Law & Economics, 50 (October): 45-79.

Berlyne, Daniel E. 1970. Novelty, Complexity and Hedonic Value. Perception and Psychophysics, 8: 279-286.

Bucks, Brian K., Arthur B. Kennickell, and Kevin B. Moore. 2006. Recent Changes in U.S. Family Finances: Evidence from the 2001 and 2004 Survey of Consumer Finances. Federal Reserve Bulletin, 92: Al-A38.

Bucks, Brian K., Arthur B. Kennickell, Traci L. Mach, and Kevin B. Moore. 2009. Changes in U.S. Family Finances from 2004 to 2007: Evidence from the Survey of Consumer Finances. Federal Reserve Bulletin, 95: A1-A55.

Caissey, G. 1990. Research, Thinking, Decision Making, Problem Solving: Skills for the Information Age. Education Digest, 55 (February): 51-53.

Cao, Henry H., Bing Han, Harold H. Zhang, and David A. Hirshleifer. 2007. Fear of the Unknown: Familiarity and Economic Decisions. Working Paper. University of North Carolina at Chapel Hill. http://ssrn.com/abstract=985381.

Clark, Robert, Gordon Goodfellow, Sylvester Schieber, and Drew Warwick. 2000. Making the Most of 401(k) Plans: Who's Choosing What and Why. In Forecasting Retirement Needs and Retirement Wealth, edited by Olivia Mitchell, Brett Hammond, and Anna Rappaport, 95-138. Philadelphia: University of Pennsylvania Press.

Coleman, Waynem, William E. Warren, and Richard Huston. 1995. Perceived Risk and the Information Search Process in the Selection of a New Dentist: An Empirical Investigation. Health Marketing Quarterly, 13 (Winter): 27-42.

Coval, Joshua D. and Tobias J. Moskowitz. 1999. Home Bias at Home: Local Equity Preference in Domestic Portfolios. Journal of Finance, 54 (December): 2045-2074.

D'Antona, John Jr. 2008. Participants Roll the Dice on Company Stock. Pensions & Investments, 36 (November): 1-2.

Fox, Craig R. and Amos Tversky. 1995. Ambiguity Aversion and Comparative Ignorance. Quarterly Journal of Economics, 110 (August): 585-603.

French, Kenneth and James Poterba. 1991. Investor Diversification and International Equity Markets. American Economic Review, 81 (May): 222-226.

Friend, Irwin and Marshall E. Blume. 1975. The Demand for Risky Assets. American Economic Review, 65 (December): 900-922.

Gourville, John and Dilip Soman. 2005. Overchoice and Assortment Type: When and why Variety Backfires. Marketing Science, 24 (Summer): 382-395.

Grable, John E., Joo-Yung Park, and So-Hyun Joo. 2009. Explaining Financial Management Behavior for Koreans Living in the United States. Journal of Consumer Affairs, 43 (Spring): 80-107.

Graham, John R., Campbell R. Harvey, and Hai Huang. 2009. Investor Competence, Trading Frequency, and Home Bias. Management Science, 55 (July): 1094-1106.

Grinblatt, Mark and Matti Keloharju. 2001. How Distance, Language, and Culture Influence Stockholdings and Trades. Journal of Finance, 56 (June): 1053-1073.

Heath, Chip and Amos Tversky. 199 I. Preference and Belief: Ambiguity and Competence in Choice under Uncertainty. Journal of Risk and Uncertainty, 4 (January): 5-28.

Hershey, Douglas A. and D. Walsh. 2000. Knowledge Versus Experience in Financial Problem Solving Performance. Current Psychology, 19 (Winter): 261-292.

Huberman, Gur. 2001. Familiarity Breeds Investment. Review of Financial Studies, 14 (Fall): 659-680.

Ivkovic, Zoran and Scott Weisbenner. 2005. Local Does as Local Is: Information Content of the Geography of Individual Investors' Common Stock Investments. Journal of Finance, 60 (February): 267-306.

Kozup, John and Jeanne M. Hogarth. 2008. Financial Literacy, Public Policy, and Consumers' Self-Protection--More Questions, Fewer Answers. Journal of Consumer Affairs, 42 (Summer): 127-136.

Laise, Eleanor. 2009. Despite Risks, Workers Guzzle Company Stock-Participants in 401(k) Plans See Safety in Employer Shares; No Other Options for a Match. Wall Street Journal, March 5. http://online.wsj.com/article/SB123621372710035183.html.

Lee, Jinkook and Jinsook Cho. 2005. Consumers' Use of Information Intermediaries and the Impact on Their Information Search Behavior in the Financial Market. Journal of Consumer Affairs, 39 (Summer): 95-120.

Liang, Nellie and Scott Weisbenner. 2002. Investor Behavior and the Purchase of Company Stock in 401(k) Plans-The Importance of Plan Design. Working Paper. University of Illinois at Urbana-Champaign. http://papers.ssrn.com/so13/papers.cfm?abstract_id=327157.

Lindamood, Suzanne, Sherman D. Hanna, and Lan Bi. 2007. Using the Survey of Consumer Finances: Methodological Considerations and Issues. Journal of Consumer Affairs, 41 (Winter): 195-222.

Massa, Massimo and Andrei Simonov. 2006. Hedging, Familiarity and Portfolio Choice. Review of Financial Studies, 19 (Summer): 633-685.

Meulbroek, Lisa K. 2005. Company Stock in Pension Plans: How Costly Is It? Journal of Law & Economics, 48 (October): 443-474.

Mitchell, Olivia and Stephen P. Utkus. 2004. The Role of Company Stock in Defined Contribution Plans. In The Pension Challenge: Risk Transfers and Retirement Income Security, edited by Olivia Mitchell and Kent Smetters, 33-70. Oxford: Oxford University Press.

Montalto, Catherine P. and Jaimie Sung. 1996. Multiple Imputations in the 1992 Survey of Consumer Finances. Financial Counseling and Planning, 7: 133-146.

Obsfeld, Maurice and Kenneth Rogoff. 2000. The Six Major Puzzles in International Macroeconomics: Is There a Common Cause? NBER Macroeconomics Annual 2000, 15: 339-390.

Perry, Vanessa G. and Marlene D. Morris. 2005. Who Is in Control? The Role of Self-Perception, Knowledge, and Income in Explaining Consumer Financial Behavior. Journal of Consumer Affairs, 39 (Winter): 299-313.

Poterba, James M. 2003. Employer Stock and 401(k) Plans. American Economic Review, 93 (May): 398-404.

Rappa, Michael. 2003. Case Study: My Simon. Managing the Digital Enterprise. http://www. startupjunkies.org/business_models.pdf.

Ritov, liana and Jonathan Baron. 1992. Status-Quo and Omission Biases. Journal of Risk & Uncertainty, 5 (February): 49-61.

Rotfeld, Herbert Jack. 2008. Financial Aliteracy. Journal of Consumer Affairs, 42 (Summer): 306-309.

Samuelson, William and Richard Zeckhauser. 1988. Status Quo Bias in Decision Making. Journal of Risk and Uncertainty, 1 (March): 7-59.

Stang, David J. 1973. Six Theories of Repeated Exposure and Affect. Catalogue of Selected Documents in Psychology, 3: 126.

U.S. Census Bureau. 2008. http://www.census.gov/population/www/socdemo/educ-attn.html.

Van Nieuwerburgh, Stijn and Laura Veldkamp. 2005. Information Immobility and the Home Bias Puzzle. NYU Working Paper FIN-04-026. http://ssrn.com/abstract=1294476.

VanDerhei, Jack, Russell Galer, Carol Quick, and John Rea. 1999. 401(k) Plan Asset Allocation, Account Balances, and Loan Activity. EBRI Issue Brief, 205.

Vanguard Group. 2007. How America Saves 2007, A Report on Vanguard 2006 Defined Contribution Plan Data. http://global.vanguard.com/international/common/pdf/CRR_HAS_2007.pdf.

Zajonc, Robert B. 1968. Attitudinal Effects of Mere Exposure. Journal of Personality and Social Psychology Monograph Supplement, 9: 1-27.

Zhu, Ning. 2003. The Local Bias of Individual Investors. Working Paper. Yale School of Management. http://faculty.gsm.ucdavis.edu/~nzhu/papers/localbias.pdf.

(1.) The Pension Protection Act of 2006 (PPA) provides roles on diversifying out of company stock. These diversification requirements apply only to plans with publicly traded company stock as an investment option and generally apply to plan years beginning after 2006. The PPA requires that plan sponsors allow participants to diversify out of company stock after three years of service for employer nonelective or matching contributions; for employee own contributions, diversification out of company stock must be permitted immediately. In addition, sponsors are required to offer at least three other materially different investment options with diversified risk and return characteristics.

Christine W. Lai (lai@saturn.yzu.edu.tw) is an Associate Professor of Finance in the College of Management, Yuan Ze University, Taiwan. Jing Jian Xiao (jfei.editor@gmail.com) is a Professor of Consumer Economics in the Department of Human Development and Family Studies at the University of Rhode Island. The authors would like to acknowledge the grant to Christine Lai from National Science Council in Taiwan (95-2745-H-155-003-HPU).
TABLE 1
Characteristics of Respondents Who Hold Positive Amounts of Company
Stock in Regular Equity Accounts

Variables                                     Mean    Frequency (%)

Company stocks/equities                      51.86%
  0 < Company stocks/equities                             22.65
   [less than or equal to] 10%
  10% < Company stocks/equities                           10.89
   [less than or equal to] 20%
  20% < Company stocks/equities                            7.55
   [less than or equal to] 30%
  30% < Company stocks/equities                            5.79
   [less than or equal to] 40%
  40% < Company stocks/equities                            6.60
   [less than or equal to] 50%
  50% < Company stocks/equities                            4.02
   [less than or equal to] 60%
  60% < Company stocks/equities                            2.95
   [less than or equal to] 70%
  70% < Company stocks/equities                            2.88
   [less than or equal to] 80%
  80% < Company stocks/equities                            3.31
   [less than or equal to] 90%
  90% < Company stocks/equities                           33.36
   [less than or equal to] 100%
Risk tolerance
    No risk                                  11.95%
    Average risk                             49.14%
    Above average risk                       34.05%
    Substantial risk                          4.86%
Consumer bias variables
    Home bias (does not own non-U.S.         91.75%
      stock)
    Company stock in DC plans                 7.06%
Competence variables
    College (highest education in college    25.69%
      but no postbachelor work)
    Graduate (highest education in           56.74%
      postbachelor graduate work)
    Head's total earnings (2007$) from job   59.49%
      is in top 25%
    Other stock/financial assets              9.11%
    Seek professional advice                 31.92%
Control variables
    Plan (employer offers stock option       32.73%
      plan)
    Have received an inheritance or gift     25.82%
Work-related characteristics
    Large firm size (employer has more       66.47%
      than 499 employees)
    Work year (number of years head has        9.98
      worked for employer)
Demographic characteristics
    Respondent Black or Hispanic              9.51%
    Male respondent                          56.06%
    Couple household                         74.45%
n = 653

Variables                                      SD       SEM

Company stocks/equities                      15.69%   0.3781%
  0 < Company stocks/equities
   [less than or equal to] 10%
  10% < Company stocks/equities
   [less than or equal to] 20%
  20% < Company stocks/equities
   [less than or equal to] 30%
  30% < Company stocks/equities
   [less than or equal to] 40%
  40% < Company stocks/equities
   [less than or equal to] 50%
  50% < Company stocks/equities
   [less than or equal to] 60%
  60% < Company stocks/equities
   [less than or equal to] 70%
  70% < Company stocks/equities
   [less than or equal to] 80%
  80% < Company stocks/equities
   [less than or equal to] 90%
  90% < Company stocks/equities
   [less than or equal to] 100%
Risk tolerance
    No risk                                  12.78%   0.1109%
    Average risk                             19.70%   0.5630%
    Above average risk                       18.68%   0.2559%
    Substantial risk                          8.47%   0.0518%
Consumer bias variables
    Home bias (does not own non-U.S.         10.84%   0.6482%
      stock)
    Company stock in DC plans                 8.23%   0.2283%
Competence variables
    College (highest education in college    17.22%   0.2076%
      but no postbachelor work)
    Graduate (highest education in           19.53%   0.4212%
      postbachelor graduate work)
    Head's total earnings (2007$) from job   19.35%   0.4707%
      is in top 25%
    Other stock/financial assets              6.11%   0.3115%
    Seek professional advice                 18.37%   0.3312%
Control variables
    Plan (employer offers stock option       18.49%   0.2790%
      plan)
    Have received an inheritance or gift     17.25%   0.2521%
Work-related characteristics
    Large firm size (employer has more       18.61%   0.3803%
      than 499 employees)
    Work year (number of years head has        4.01    0.1043
      worked for employer)
Demographic characteristics
    Respondent Black or Hispanic             11.56%   0.1729%
    Male respondent                          19.56%   0.6834%
    Couple household                         17.19%   0.6967%
n = 653

Note: Author calculation based on the weighted sample of 2004
and 2007 SCFs.

TABLE 2
Risk Tolerance Attitudes and Investment Behaviors

Panel A: Summary Statistics (Weighted) (n = 653)

                                    Equities/Financial Assets
                              Mean (%)       SD (%)        SEM (%)
Willing to take
  No risk                       10.45         9.65         0.6980
  Average risk                  24.41         10.8         0.4335
  Above average risk            35.99        10.62         0.2565
  Substantial risk              35.36         8.63         0.8039

                                 Company Stock/Financial Assets
                               Mean (%)      SD (%)        SEM (%)
Willing to take
  No risk                       39.75        16.57         1.3270
  Average risk                  22.52        10.12         0.2162
  Above average risk            20.24         8.75         0.2890
  Substantial risk              27.35         8.02         1.0328

Panel B: OLS regression (n = 653) (based on five implicates with RII)

                                    Equities/Financial Assets
Model                              Coefficient     t-Statistics

Intercept                           0.4018 ***       24.3667
No risk                            -0.2742 ***       -6.7023
Average risk                       -0.0837 ***       -3.7355
Adjusted R-squared (average         0.0694
  of five implicates)

                                   Company Stock/Financial Assets
Model                               Coefficient     t-Statistics

Intercept                           0.3698 ***       16.7844
No risk                             0.3562 ***        6.5581
Average risk                        0.0683 *          2.2821
Adjusted R-squared (average         0.0600
  of five implicates)

* p < .05, ** p < .01, *** p < .001.

TABLE 3
Results of ANOVA (Based on Five Implicates with RII) (Dependent
Variable: Company Stock in Regular Equity Accounts/Equity in
Regular Equity Accounts)

Independent Variables                     Mean (%)   F-Value   p-Value

  Risk tolerance                                      72.93    0.0000
  No risk                                  72.60
  Average risk                             43.80
  Above average risk                       36.78

  Substantial risk                         37.77

Consumer biases
Home bias (does not own non-U.S. stock)              292.34    0.0000
  Home country stock                       47.50
  Other country stock                      16.78

Mere exposure effect (dummies)
  Company stock in DC plans                           12.78    0.0000
    Bottom third of the value              43.08
    Middle third                           51.08
    Top third                              63.98
Consumer competences
Education (dummies)                                  198.61    0.0000
  Lower than college                       68.40
  College (highest education in college    58.47
    but no postbachelor work)
  Graduate (highest education in           35.72
    postbachelor graduate work)
Income (dummy)                                       140.85    0.0000
  Head's total earnings (2007$) from       54.92
    job is in top 25% = no
  Head's total earnings (2007$) from       38.25
    job is in top 25% = yes
Otherstock/financial assets (dummies)                803.95    0.0000
  Bottom third of the value                70.46
  Middle third                             32.81
  Top third                                21.83
Seek professional advice (dummy)                      46.18    0.0000
  Yes                                      36.99
  No                                       46.31

Independent Variables                             Comparisons

  Risk tolerance
  No risk
  Average risk                            No risk vs. avg. risk
  Above average risk                      Avg. risk vs. above
                                           avg. risk
  Substantial risk                        Above avg. risk vs.
                                           substantial risk
Consumer biases
Home bias (does not own non-U.S. stock)
  Home country stock
  Other country stock                     Home vs. other country
                                           stock
Mere exposure effect (dummies)
  Company stock in DC plans
    Bottom third of the value
    Middle third                          Top vs. bottom third
    Top third                             Top vs. middle third
Consumer competences
Education (dummies)
  Lower than college
  College (highest education in college   College vs. lower than
    but no postbachelor work)              college
  Graduate (highest education in          Graduate vs. college
    postbachelor graduate work)
Income (dummy)
  Head's total earnings (2007$) from
    job is in top 25% = no
  Head's total earnings (2007$) from      Income (top 25%) vs. Income
    job is in top 25% = yes                (not in top 25%)
Otherstock/financial assets (dummies)
  Bottom third of the value
  Middle third                            Middle vs. bottom third
  Top third                               Top vs. middle third
Seek professional advice (dummy)
  Yes                                     Seek professional advice
  No                                      Yes vs. No

Independent Variables                     Mean Difference (%) (a)

  Risk tolerance
  No risk
  Average risk                                   28.80 ***
  Above average risk                              7.02 ***

  Substantial risk                                 -0.99

Consumer biases
Home bias (does not own non-U.S. stock)
  Home country stock
  Other country stock                            30.72 ***

Mere exposure effect (dummies)
  Company stock in DC plans
    Bottom third of the value
    Middle third                                 20.90 ***
    Top third                                     12.90 **
Consumer competences
Education (dummies)
  Lower than college
  College (highest education in college          -9.93 ***
    but no postbachelor work)
  Graduate (highest education in                -22.75 ***
    postbachelor graduate work)
Income (dummy)
  Head's total earnings (2007$) from
    job is in top 25% = no
  Head's total earnings (2007$) from            -16.67 ***
    job is in top 25% = yes
Otherstock/financial assets (dummies)
  Bottom third of the value
  Middle third                                  -37.65 ***
  Top third                                     -10.98 ***
Seek professional advice (dummy)
  Yes                                            -9.32 ***
  No

(a) Significance in mean difference.

* p < .05, ** p < .01, *** p < .001.

TABLE 4
Regression Analysis (Based on Five Implicates with RII) (n = 653)
(Dependent Variable: Company Stock in Regular Equity Accounts/
Equity in Regular Equity Accounts)Independent Variables

                                                         Coefficient
Intercept                                                     0.5870
Risk tolerance (omitted category: Above
average risk + Substantial risk)
  No risk                                                     0.1915
  Average risk                                                0.0294
Consumer biases
Home bias
  Home bias (does not own non-U.S. stock)                     0.1184
Mere exposure effect
  Company stock in DC plans                                   0.2160
Consumer competences
Education (omitted category: Lower than college)
  College (highest education in college but no post           -0.085
    bachelor work)
  Graduate (highest education in postbachelor graduate       -0.1921
    work)
Income
  Head's total earnings (2007$) from job is in top 25%       -0.0967
Other stock/financial assets
  Other stock/financial assets                               -0.6716
Seek professional advice
  Yes                                                        -0.0942
Control variables
  Plan (employer offers stock option plan)                    0.0714
  Have received an inheritance or gift                       -0.0598
Other work-related and demographic variables
Large firm size (employer has more than 499 employees)        0.0393
Work year (number of years head has worked for                0.0005
  employer)
Respondent Black or Hispanic                                  0.0378
Male respondent                                              -0.0122
Couple household                                             -0.0006
Year dummy (2007 = 1)                                         0.0387
Adjusted R-squared (average of five implicates)               0.3267
F                                                            18.4328
p                                                           <0.0001

                                                         t-Statistics
Intercept                                                      8.4141
Risk tolerance (omitted category: Above
average risk + Substantial risk)
  No risk                                                      3.9179
  Average risk                                                 1.1237
Consumer biases
Home bias
  Home bias (does not own non-U.S. stock)                      2.8901
Mere exposure effect
  Company stock in DC plans                                    2.7098
Consumer competences
Education (omitted category: Lower than college)
  College (highest education in college but no post           -1.7876
    bachelor work)
  Graduate (highest education in postbachelor graduate        -4.5539
    work)
Income
  Head's total earnings (2007$) from job is in top 25%        -3.1477
Other stock/financial assets
  Other stock/financial assets                                -9.0444
Seek professional advice
  Yes                                                         -3.5035
Control variables
  Plan (employer offers stock option plan)                     2.3481
  Have received an inheritance or gift                         -2.145
Other work-related and demographic variables
Large firm size (employer has more than 499 employees)         1.3908
Work year (number of years head has worked for                 0.4239
  employer)
Respondent Black or Hispanic                                   0.7300
Male respondent                                               -0.4266
Couple household                                              -0.0173
Year dummy (2007 = 1)                                          1.5487
Adjusted R-squared (average of five implicates)
F
p

                                                         p-Value
Intercept                                                 0.0000
Risk tolerance (omitted category: Above
average risk + Substantial risk)
  No risk                                                 0.0001
  Average risk                                            0.2613
Consumer biases
Home bias
  Home bias (does not own non-U.S. stock)                 0.0045
Mere exposure effect
  Company stock in DC plans                               0.0075
Consumer competences
Education (omitted category: Lower than college)
  College (highest education in college but no post       0.0739
    bachelor work)
  Graduate (highest education in postbachelor graduate    0.0000
    work)
Income
  Head's total earnings (2007$) from job is in top 25%    0.0017
Other stock/financial assets
  Other stock/financial assets                            0.0000
Seek professional advice
  Yes                                                     0.0005
Control variables
  Plan (employer offers stock option plan)                0.0190
  Have received an inheritance or gift                    0.0321
Other work-related and demographic variables
Large firm size (employer has more than 499 employees)    0.1649
Work year (number of years head has worked for            0.6717
  employer)
Respondent Black or Hispanic                              0.4654
Male respondent                                           0.6698
Couple household                                          0.9862
Year dummy (2007 = 1)                                     0.1216
Adjusted R-squared (average of five implicates)
F
p

TABLE 5
Hierarchical Regressions (n = 653) (Dependent Variable: Company
Stock in Regular Equity Accounts/Equity in Regular Equity Accounts)

                                          Model 1

Variables                          Coefficient   p-Value

Controls
  Intercept                             0.5480    0.0000
  Plan (employer offers stock           0.0607    0.0767
    option plan)
  Have received an inheritance         -0.1008    0.0015
    or gift
  Large firm size (employer has         0.0593    0.0644
    more than 499 employees)
  Work year (number of years           -0.0008    0.5261
    head has worked for
    employer)
  Respondent Black or Hispanic          0.1785    0.0020
  Male respondent                      -0.1068    0.0005
  Couple household                     -0.0836    0.0256
  Year dummy                            0.0018    0.9497
  [R.sup.2] [DELTA]                     0.0777
Risk tolerance (omitted category:
 Above average risk +
 Substantial risk)
  No risk
  Average risk
Consumer biases
  Home bias (does not own
  Don-U.S. stock)
  Company stock in DC plans
  [R.sup.2] [DELTA]
  F-test ([F.sub.(4,641)])
Consumer competences
  College (highest education
    incollege but no post
    bachelor work)
  Graduate (highest education
    in postbachelor
    graduate work)
  Head's total earnings (2007$)
    from job is in top 25%
  Other stock/financial assets
Seek professional advice = yes
  [R.sup.2]0
  F-test ([F.sub.(5,636)])
Total [R.sup.2]                         0.0777

                                          Model 2

Variables                          Coefficient   p-Value

Controls
  Intercept                             0.2333    0.0001
  Plan (employer offers stock           0.0582    0.0754
    option plan)
  Have received an inheritance         -0.0751    0.0135
    or gift
  Large firm size (employer has         0.0364    0.2405
    more than 499 employees)
  Work year (number of years           -0.0009    0.4647
    head has worked for
    employer)
  Respondent Black or Hispanic          0.1284    0.0206
  Male respondent                      -0.0585    0.0520
  Couple household                     -0.0539    0.1344
  Year dummy                            0.0121    0.6578
  [R.sup.2] [DELTA]
Risk tolerance (omitted category:
 Above average risk +
 Substantial risk)
  No risk                               0.2879    0.0000
  Average risk                          0.0519    0.0662
Consumer biases
  Home bias (does not own               0.2395    0.0000
  Don-U.S. stock)
  Company stock in DC plans             0.1996    0.0389
  [R.sup.2] [DELTA]                     0.0970
  F-test ([F.sub.(4,641)])             18.8260    0.0000
Consumer competences
  College (highest education
    incollege but no post
    bachelor work)
  Graduate (highest education
    in postbachelor
    graduate work)
  Head's total earnings (2007$)
    from job is in top 25%
  Other stock/financial assets
Seek professional advice = yes
  [R.sup.2]0
  F-test ([F.sub.(5,636)])
Total [R.sup.2]                         0.1747

                                          Model 3

Variables                          Coefficient   p-Value

Controls
  Intercept                             0.5870    0.0000
  Plan (employer offers stock           0.0714    0.0190
    option plan)
  Have received an inheritance         -0.0598    0.0321
    or gift
  Large firm size (employer has         0.0393    0.1649
    more than 499 employees)
  Work year (number of years            0.0005    0.6717
    head has worked for
    employer)
  Respondent Black or Hispanic          0.0378    0.4654
  Male respondent                      -0.0122    0.6698
  Couple household                     -0.0006    0.9862
  Year dummy                            0.0387    0.1216
  [R.sup.2] [DELTA]
Risk tolerance (omitted category:
 Above average risk +
 Substantial risk)
  No risk                               0.1915    0.0001
  Average risk                          0.0294    0.2613
Consumer biases
  Home bias (does not own               0.1184    0.0045
  Don-U.S. stock)
  Company stock in DC plans             0.2160    0.0075
  [R.sup.2] [DELTA]
  F-test ([F.sub.(4,641)])
Consumer competences
  College (highest education            -0.085    0.0739
    incollege but no post
    bachelor work)
  Graduate (highest education          -0.1921    0.0000
    in postbachelor
    graduate work)
  Head's total earnings (2007$)        -0.0967    0.0017
    from job is in top 25%
  Other stock/financial assets         -0.6716    0.0000
Seek professional advice = yes         -0.0942    0.0005
  [R.sup.2]0                            0.1520
  F-test ([F.sub.(5,636)])             28.7234    0.0000
Total [R.sup.2]                         0.3267
COPYRIGHT 2010 American Council on Consumer Interests
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2010 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Lai, Christine W.; Xiao, Jing Jian
Publication:Journal of Consumer Affairs
Article Type:Report
Geographic Code:1USA
Date:Mar 22, 2010
Words:12727
Previous Article:Concern with immediate consequences magnifies the impact of compulsive buying tendencies on college students' credit card debt.
Next Article:Cohort effects of household expenditures on food away from home.
Topics:

Terms of use | Privacy policy | Copyright © 2019 Farlex, Inc. | Feedback | For webmasters