Do financial networks matter in retirement investment
decisions? Evidence from Generation Yers

Yunhyung Chunga, Youngkyun Parka,*
aCollege of Business and Economics, University of Idaho, 875 Perimeter Drive MS 3161,

Moscow, ID 83844-3161, USA

Abstract

Using experimental survey data collected from a sample of Generation Yers, we examine the joint
influence of financial literacy and financial networks on individual retirement investment decisions.
We find, first, that financial literacy and financial network intensity (the network strength with the
financially literate) are positively related to stock allocation. Second, the positive relationship between
financial literacy and stock allocation, however, is significant only among those having high financial
network intensity. This finding suggests that the positive effects of financial literacy documented in the
literature can be limited to only those who have strong networks with the financially literate. © 2015
Academy of Financial Services. All rights reserved.

Jel classification: G11; J26

Keywords: Retirement investment decisions; Financial networks; Financial literacy

1. Introduction

While the financial environment becomes complex and volatile, individual financial
decision-making has been considered an important factor of determining an individual’s
financial well-being. Thus, a large body of research has investigated what factors influence
financial decision-making and has found that financial literacy is a key determinant in many
areas such as money management, credit, investment, and retirement planning (e.g., Camp-
bell, 2006; Lusardi & Tufano, 2009; Moore, 2003; Perry & Morris, 2005). However,

* Corresponding author. Tel.: �1-208-885-7154; fax: �1-208-885-5347.
E-mail address: youngpark@uidaho.edu (Y. Park)

Financial Services Review 24 (2015) 77–99

1057-0810/15/$ – see front matter © 2015 Academy of Financial Services. All rights reserved.



financial literacy alone may not be able to improve financial behaviors substantially (e.g.,
Fernandes, Lynch, & Netemeyer, 2014; Kiviat & Morduch, 2012).1 Because of the fast-
changing financial product offerings and industry environment, acquiring and processing
up-to-date financial knowledge and information is quite challenging even to those who are
financially literate (Willis, 2011). Consequently, individual investors often rely on other
people to make investment decisions, instead of solely on their own knowledge.

Relatively recently, research on financial decision-making has begun to recognize the
importance of social interactions in individual investment decisions. Because individual
investors may be able to reduce time and effort to acquire financial knowledge and infor-
mation through social interactions with others, “social” people are more likely to participate
in financial markets than others who do not engage in social interactions (Brown, Ivković,
Smith, & Weisbenner, 2008; Duflo & Saez, 2002; Hong, Kubik, & Stein, 2004; Ivković &
Weisbenner, 2007; Kaustia & Knupfer, 2012; Lu, 2011). In line with this argument, research
on social networks in management finds that individuals may acquire a high volume of and
diverse work-related knowledge and information through frequent communication with
co-workers or experts who have task-related knowledge (Hansen, Mors, & Lovas, 2005;
McFadyen & Cannella, 2004; Reinholt, Pedersen, & Foss, 2011; Wang & Noe, 2010).
Because tacit and confidential knowledge tend to be shared only with whom people trust
(Nahapiet & Goshal, 1998; Yli-Renko, Autio, & Sapienza, 2001) and transferred through
strong network ties (Chung & Jackson, 2013; Hansen, Podolny, & Pfeffer, 2001), strong
connections with people who are financially literate may facilitate individuals to obtain
reliable and opportune information for investments without excessive time and effort.

Drawing on the literature on financial literacy and social networks, we propose that an
individual investor’s decision-making may be affected by not only his or her financial
knowledge but also social networks. Among various types of social networks (e.g., career
advice networks and task information networks), we focus on individuals’ social networks
for acquiring financial or investment information, which are dubbed “financial networks.”
While prior studies focused on social activities, few studies examined how social networks
influence investment decisions.2 Thus, this study aims to fill this gap by taking account of
financial networks in discussing individuals’ investment decisions. Among financial network
characteristics, we focus on financial network intensity, defined as communication frequency
with the financially literate, because frequent interactions with the financially literate may
enable an individual investor to acquire critical and reliable information and knowledge for
his or her investments.

Taken together, we examine how financial knowledge and financial network intensity are
associated with individual investment decisions. Among various individual investment
decisions, we consider the following sequential decision-making process for retirement
investment: first, whether an individual investor chooses a default investment option or not
and, second, when selecting “no default,” to what extent he/she allocates contributions to
stocks. For this examination, we collected experimental survey data from senior business
college students at age 20–26 in 2012, who belong to the Generation Y (ages 18–35 in
2012). Most of them will enter an early stage of their working career upon graduation and
likely become defined contribution plan participants. We chose this sample because retire-
ment investment decisions in their early career may significantly affect their retirement

78 Y. Chung, Y. Park / Financial Services Review 24 (2015) 77–99



wealth. Once new employees make a retirement investment decision, they tend to stick with
the status quo and avoid changing their retirement investment portfolio (Choi, Labibson,
Madrian, & Metrick, 2004, 2006; Madrian & Shea, 2001). In addition, given that the Social
Security trust fund is projected to be exhausted in 2034 and at that time that Social Security
will be able to cover only 75% of scheduled benefits (Social Security and Medicare Boards
of Trustees, 2014), how to construct a retirement portfolio would be more important to
Generation Yers than older generations.

With a sample (N � 97) of senior students in a business college, first, a choice of no
default in a retirement plan is positively associated with financial literacy (or financial
knowledge), but not significantly associated with financial network intensity (or the network
strength with the financially literate).3 Second, respondents’ contribution allocation to stock
funds is positively associated with financial literacy or financial network intensity. Third,
financial literacy does not interact with financial network intensity on a choice of a default
option, but they interact with each other on respondents’ contribution allocation to stock
funds in a retirement portfolio. In an additional analysis on the interaction effect, the positive
effects of financial literacy on stock allocation in a retirement portfolio are significant only
among those who have high financial network intensity.4 The interaction between financial
literacy and financial network intensity also affects how much respondents’ retirement
portfolios deviate from age-appropriate stock allocations, the deviation that is evaluated with
respect to the Morningstar Lifetime Allocation Indexes. We find that those with a high level
of financial literacy have smaller deviations from the age-appropriate stock allocations only
when they have strong networks with the financially literate.

This study enhances an understanding of retirement investment decision-making and
long-term financial planning. While extant research finds financial literacy as a key deter-
minant in retirement investment decision-making and long-term financial planning (e.g.,
Alhenawi & Elkhal, 2013; Lusardi & Mitchell, 2007a, 2007b), our research finds that
financial literacy and social networks interplay in retirement investment decision-making.
Specifically, the positive effects of financial literacy on retirement investment decisions
documented in the literature can be limited to only those who have strong networks with
financially literate people.

The results of this study may also provide practical implications to individual investors,
financial planners, and employer-sponsored financial education programs. To improve in-
vestment decisions, individual investors’ own financial knowledge alone may not be enough;
when they synthesize their own financial knowledge and the knowledge acquired through
interactions with the financially literate, their investment decisions can be significantly
enhanced. Therefore, our findings shed light on the importance of financial networks with
family, co-workers, and professional financial planners when individual investors make
investment decisions. Finally, employer-sponsored financial education programs may be
designed not only to elevate employees’ financial knowledge but also to provide more
opportunities to build networks for financial advice (such as more opportunities to commu-
nicate with financial planners or advisors).

The remainder of the article is organized as follows. Section 2 provides a literature review
on financial literacy, financial network intensity, and retirement investment decisions and
addresses hypotheses. Section 3 describes the sample, survey instruments, and model

79Y. Chung, Y. Park / Financial Services Review 24 (2015) 77–99



specification to test the hypotheses. Section 4 presents regression results indicating that
financial literacy and/or financial network intensity can affect retirement investment deci-
sions. Section 5 concludes with a summary of findings and suggestions for future research.

2. Financial literacy, financial network intensity, and retirement investment
decisions

Managing an investment portfolio requires an individual investor to spend significant time
and effort in understanding various finance concepts such as asset returns, volatility, and
covariance between asset returns. Thus, information costs—for example, costs of acquiring
and processing information about risks and returns—represent a significant barrier for
individual investors to build their own investment portfolio (Christelis, Jappelli, & Padula,
2010; Vissing-Jørgensen, 2003). Building on the notion of information cost, we argue that
individual investors’ financial literacy and financial networks reduce information cost, which
consequently influences their retirement investment portfolios.

2.1. Financial literacy

Information costs for investments may be substantially large for people who have low
financial knowledge and skills (Vissing-Jørgensen, 2003). Because they have to spend
significant time and effort in acquiring and processing financial information, they may give
up the costly process for investment decision-making and seek a simple solution (Bettman
& Park, 1980). Agnew and Szykman (2005) find that individual investors with a low level
of financial knowledge are likely to opt for a default investment option more often than those
with a high level of financial knowledge. In addition, Choi et al. (2004) find from several
401(k) plans that young, female, and low-income participants with short tenure, who are
typically associated with a low level of financial literacy (Banerjee, 2011; Hung, Parker, &
Yoong, 2009; Lusardi & Mitchell, 2007a), are more likely to choose a default investment
option than others. Based on the theoretical and empirical findings in the previous literature,
we argue that individuals who have little knowledge of finance are more likely to select a
default investment option than those who have much knowledge of finance. On the contrary,
individuals with a high level of financial literacy are more likely to choose no default and
build their own retirement investment portfolio than those with a low level of financial
literacy. In this respect, we hypothesize the following:

H1a: Financial literacy is positively associated with a choice of “no default” in a
retirement plan.

Individual investors who are financially literate may be able to reduce the costs of
acquiring and processing financial information and, as a result, increase stockholding.5

Because financial literacy helps reduce the fixed costs of acquiring and processing financial

80 Y. Chung, Y. Park / Financial Services Review 24 (2015) 77–99



information (e.g., Vissing-Jørgensen, 2003),6 financially literate investors are likely to hold
stocks (van Rooij et al., 2011; Yoong, 2010).7 In addition, once individual investors acquire
financial knowledge or information, they may apply this knowledge or information to a
larger volume of the risky asset with an increase in the expected rate of return (Delavande,
Rohwedder, & Willis, 2008). In this respect, we expect that individual investors with a high
level of financial literacy are likely to allocate more contributions to stocks than those with
a low level of financial literacy, all else being equal.

H1b: Financial literacy is positively associated with stock allocation in a retirement
investment portfolio.

2.2. Financial network intensity

Individuals’ investment decisions are affected by not only their own financial literacy
but also social interactions such as word-of-mouth or observational learning (Banerjee,
1992; Bikchandani et al., 1992; Ellison & Fudenberg, 1993, 1995). For example,
individuals’ investment behavior is affected by investment decisions or outcomes of their
co-workers (Duflo & Saez, 2002; Lu, 2011) or neighbors (Brown et al., 2008; Hong et
al., 2004; Ivković & Weisbenner, 2007; Kaustia & Knupfer, 2012) because social
interactions, like financial literacy, may be able to reduce the costs of acquiring and
processing financial information.

Network ties can serve both instrumental/informational and expressive/social pur-
poses (Balkundi & Harrison, 2006; Cascioaro & Lobo, 2008). This study focuses on the
instrumental role of networks, assuming that instrumental external ties are most relevant
to acquiring financial knowledge and information. For instrumental external ties, the
network strength (in terms of communication frequency) with the financial literate is of
our particular interest; the costs of acquiring and processing information related to
investment may be reduced by frequent interactions with people who are financially
knowledgeable. Social network literature documents that frequent communications gen-
erate “strong ties” and, thus, trust between people in the network (e.g., Krackhardt,
1992). Trustworthy information built on frequent communications in one’s networks,
then, can affect his or her investment decisions (e.g., Guiso, Sapienza, & Zingales,
2008).8 Because people tend to share critical and confidential knowledge only with
whom they trust (Nahapiet & Goshal, 1998; Yli-Renko, Autio, & Sapienza, 2001),
frequent interactions with people who have high financial literacy may provide individ-
uals easy access to a high quality of investment and knowledge. Therefore, through
social networks with the financially literate, individuals may acquire reliable and op-
portune information for their investments.

Drawing on the studies on social network, we expect that financial network intensity,
indicating the relationship strength with the financially literate, may influence individuals’
investment decisions. Specifically, for a given default investment option, we expect that

81Y. Chung, Y. Park / Financial Services Review 24 (2015) 77–99



individual investors who have strong networks with the financially literate are likely to build
their own retirement portfolio instead of selecting a default option. Thus, we hypothesize the
following:

H2a: Financial network intensity is positively associated with a choice of “no default”
in a retirement plan.

A strong relationship with people who have a high level of financial literacy may also
encourage individuals to increase stock allocation in their retirement investment portfolio.
Because individuals are likely to obtain critical and reliable financial or investment infor-
mation through frequent interactions with the financially literate, they may significantly be
able to reduce time and effort to obtain and process the information necessary for their
investments. Individuals usually have limited time and effort to spend on constructing a
retirement portfolio. Thus, strong networks with financially literate people (i.e., high finan-
cial network intensity) may enable individual investors to easily obtain investment informa-
tion with lower costs and, as a result, to increase stockholdings. In this regard, we hypoth-
esize the following:

H2b: Financial network intensity is positively associated with stock allocation in a
retirement investment portfolio.

2.3. Interactions of financial literacy and financial network intensity

In the previous sections, we developed the hypotheses that financial literacy or financial
network intensity affects individuals’ investment decisions for retirement. The two factors
may also interact for retirement investment decisions. Specifically, we argue that financially
knowledgeable individuals are more likely to build their own portfolio and increase stock
allocations when they have higher financial network intensity (i.e., stronger networks with
the financially literate) than otherwise. Individual investors who have high financial knowl-
edge may have greater ability to understand broad and diverse knowledge about financial
investment obtained through interactions with the financially literate. In addition, financially
knowledgeable investors may easily synthesize their own financial knowledge and the
knowledge acquired through social networks with the financially literate. As a result,
financial networks may enable individual investors with high financial literacy to reduce their
opportunity cost of obtaining and processing financial knowledge, which consequently eases
their investment decision-making. Hence, we argue that individual investors’ financial
network intensity complements their financial literacy: any positive effects of financial
literacy on retirement investment decisions would be stronger among those having high
financial network intensity than those having low financial network intensity. Accordingly,
we expect the following:

82 Y. Chung, Y. Park / Financial Services Review 24 (2015) 77–99



H3a: Financial network intensity interacts with financial literacy on the selection of a
default investment option in a retirement plan.

H3b: Financial network intensity interacts with financial literacy on stock allocation in
a retirement investment portfolio.

3. Sample, survey instruments, and model specification

3.1. Sample and procedure

To examine the effects of financial literacy and financial network intensity on investment
decisions, we collected data from 111 senior college students in a business school located in
the northwestern United States, using a traditional paper-and-pencil survey in the classroom
in Spring and Fall 2012 semesters. To enhance survey participants’ calculation accuracy,
calculators were distributed at the beginning of the survey. Students who participated in the
survey received class engagement credits. Four respondents who were 35 or older and 10
respondents who did not provide all the information are excluded from the sample, resulting
in a sample of 97 respondents.

3.2. Experimental survey design, model specification, and measures

3.1.1. Investment decisions for retirement
Using an experimental survey method, we developed a scenario regarding investment

decisions in a context of a 401(k) plan. In the scenario, we asked respondents to assume that
they were recently hired and asked to make investment decisions for a 401(k) plan with an
array of investment options provided by a company. We assessed two dependent variables
from sequential retirement investment decisions. First, survey respondents were asked to
make a decision for a default investment option, which is a money market fund. If a
respondent did not choose the default option, then he or she was asked to build his or her own
retirement portfolio from an array of investment options: 10 different mutual funds including
a money market fund.9 The information for each fund was provided in a separate brochure
so that survey participants could compare investment objectives, strategies, expenses, risk
and return, and the performance of the funds. The fund information was provided using
actual funds in the marketplace, but actual fund names including fund family names were not
disclosed in the brochure to prevent respondents from selecting a fund only because of their
familiarity with fund family names. Table 1 summarizes fund information provided for
survey participants. Respondents were asked to use the fund information and allocate their
contributions as a percentage to each fund up to the total contribution of 100%.

3.1.2. Model specification
To reflect respondents’ sequential investment decisions, we estimate the following Heck-

man two-stage selection model:

83Y. Chung, Y. Park / Financial Services Review 24 (2015) 77–99



Selection: s*i � �0 � �1FLi � �2FNi � �3�FLiFNi� � �4Zi � ui (1)

si � 1 if s*i � 0 and si � 0 if s*i � 0

Stock allocation: yi � �0 � �1FLi � �2FNi � �3�FLiFNi� � �4Wi � �i (2)

We use the Heckman two-stage selection model because the two-stage selection model
would overcome a potential selection bias problem that the error term (�i) in Eq. (2) can be
correlated with the error term (ui) in Eq. (1). In the selection equation, the dependent variable
(si) is a dummy, which indicates whether individual i chooses the default option—that is, a
money market fund—or not. If a respondent selects “no default,” the variable takes a value
of 1 and otherwise 0. In the stock allocation equation, the dependent variable (yi) indicates
individual i’s contribution allocation to stock funds as a percentage. Stock allocation (yi) is
calculated by summing respondent i’s contribution allocations to domestic and international
stock funds as a percentage. As key independent variables, the model includes FLi and FNi,
which indicate individual i’s financial literacy and financial network intensity, respectively,
and their interaction term (FLiFNi). Zi and Wi indicate a set of control variables for the
selection equation and the stock allocation equation, respectively.

3.1.3. Financial literacy
To measure an individual’s financial literacy, first, we assume that financial literacy can

be captured by a level of financial knowledge, following prior studies on financial literacy
(e.g., Alhenawi and Elkhal, 2013; Collins, 2012; Hilgert et al., 2003; Lusardi and Mitchell,

Table 1 Summary fund information provided for survey participants

Fund types
1. Money market fund (designated as a default investment option)
2. Small-cap growth fund
3. Mid-cap blend fund
4. Large-cap value fund
5. Large-cap growth fund
6. Foreign small/mid-cap blend fund
7. Foreign large-cap value fund
8. Short-term bond fund
9. Intermediate-term bond fund
10. Inflation-protected bond fund

Fund information
1. Investment objective
2. Principal investment strategies
3. Principal investment risks
4. Annual expense ratio
5. Return and risk

a. Three-year average return, SD, and Sharpe ratio
b. Five-year average return, SD, and Sharpe ratio

6. 10-year total annual returns (a bar graph including data labels)

Notes: The information for ten funds was provided in a separate brochure so that survey participants could
compare investment objectives, strategies, expenses, risk and return, and the performance of the funds. The fund
information was provided using actual funds in the marketplace, but actual fund names including fund family
names were not disclosed in the brochure to prevent respondents from selecting a fund due to family names.

84 Y. Chung, Y. Park / Financial Services Review 24 (2015) 77–99



2007a). Next, to evaluate respondents’ financial knowledge, we use a battery of five
questions from the National Financial Capability Study supported by the FINRA Investor
Education Foundation. These questions cover fundamental concepts of economics and
finance, such as calculations about interest rates and inflation, the relationship between
interest rates and bond prices, the relationship between interest payments and maturity in
mortgages, and risk diversification.10 A higher score indicates a higher level of financial
literacy.

3.1.4. Financial network intensity
Following Chung and Jackson (2013) and Reinholt, Pedersen, and Foss (2011), we use an

egocentric network technique to assess respondents’ financial network intensity. Respon-
dents list the first and last name initials of up to 10 people (including spouse/partner, parents,
siblings, relatives, friends, co-workers, financial planners or advisors, and others) who they
believe are most important sources to their financial or investment decisions. Limiting the list
of possible contacts to 10 people may not allow respondents to describe their entire financial
networks, but constraining the number of contacts listed has the benefit of making data
collection more feasible (Morrison, 2002).

For each person in their network, respondents provide scores to answer two questions:
communication frequency and a level of financial knowledge (see Table 2). Using the
responses to the two questions, we yield an individual’s financial network intensity measured
by communication frequency with the financially literate in his or her financial networks. We
define individual i’s financial network intensity (FNi) as follows:

FNi �
�j�1

J �CFREQij � NFKij�

MaxZi
for j � 1, 2, ..., J � � 10� (3)

Table 2 Summary of questions for financial networks

Question 1. During the five years, how often have you talked to each person to acquire financial and/or
investment information?
0 � Never
1 � Once a year or less
2 � Several times a year
3 � Once a month
4 � 2–3 times a month
5 � Once a week or more

Question 2. Indicate to what extent each person has financial knowledge.
1 � Very little
2 � Below average
3 � Average
4 � Above average
5 � Very high

Notes: Survey respondents were asked to list the first and last name initials of up to 10 people (including
spouse/partner, parents, siblings, relatives, friends, coworkers, financial planners or advisors, and others) who
they believe are most important to their financial or investment decisions. For each person respondents listed, they
were also asked to answer the above two questions about communication frequency and financial literacy.

85Y. Chung, Y. Park / Financial Services Review 24 (2015) 77–99



First, individual i identifies persons (j) in his or her financial network (called “alters”), but
the number of alters is constrained up to 10 (J �10). J indicates the total number of persons
in individual i’s financial networks. Second, individual i evaluates communication frequency
with a person j in his or her network (CFREQij), using a scale of 0 (never) to 5 (once a week
or more) (Table 2: Question 1). We assume that a large value of communication frequency
indicates a strong relationship with a person in one’s network. Third, individual i evaluates
the financial literacy of person j, using a scale of 0 (very little) to 5 (very high) (Table 2:
Question 2). The responses are converted to a binary variable (NFKij), which indicates
whether person j in individual i’s network is financially literate or not. NFKij takes a value
of one for a response of “above average” or “very high” and a value of zero for a response
of “very little,” “below average,” or “average.” Fourth, because NFKij is zero for person j’s
average or lower financial literacy in individual i’s network, the numerator in the Eq. (3) is
computed by summing up individual i’s communication frequencies with only the financially
literate in his or her network. Finally, individual i’s network strength with the financially
literate is evaluated with respect to the maximum possible value (MaxZi) that the individual
can have from the relationships with the financially literate. MaxZi is defined as a product of
J, the maximum possible value of CFREQij (� 5), and the maximum possible value of NFKij

(� 1) for individual i. For example, when individual i has five persons in his or her financial
networks, his or her maximum possible value (MaxZi) is 25 (� 5 � 5 � 1). Thus, individual
i’s financial network intensity (FNi) takes a value between 0 and 1; a larger value of FNi

indicates greater network strength with the financially literate.

3.1.5. Control variables
Respondents’ network size, investment experience, risk tolerance, grade point average

(GPA), current mood state, and demographic backgrounds are included as control variables.
First, following Hansen et al. (2001), we define network size as the total number of

persons from whom a respondent acquires financial or investment information. Those who
have a larger size of network may have more chances to obtain a broader spectrum of
information and knowledge (e.g., Chung and Jackson, 2013; Gargiulo, Ertug, and Galunic,
2009).

Second, we include respondents’ risk tolerance, following prior studies on financial risk
tolerance and investment behavior (e.g., Gibson, Michayluk, and Venter, 2013; Guiso et al.,
2008; Hong et al., 2004; van Rooij et al., 2011; Yoong, 2010). Respondents’ decisions on a
choice of a default option and stock allocation may be affected by their risk tolerance. To
measure risk tolerance, we use a battery of 13 items developed by Grable and Lytton
(1999).11

Third, we control for investment experience. We expect that a respondent’s prior invest-
ment experience may influence his or her decisions on a choice of a default option and stock
allocation. Thus, survey participants were asked to answer whether they have owned any
stocks or mutual funds during the past five years. When a respondent answers “Yes” in either
stocks or mutual funds, the respondent is regarded as one who has investment experience.

Fourth, we control for the current mood state by including the 20-item Positive Affect and
Negative Affect Scale (PANAS; Watson et al., 1988), relying on the findings of the literature
on personality and social psychology that risk perceptions and associated choices are

86 Y. Chung, Y. Park / Financial Services Review 24 (2015) 77–99



significantly influenced by the mood state (or feelings) at the moment of decision making
(e.g., Forgas, 1995; Isen, 2000; Leith and Baumeister, 1996; Loewenstein et al., 2001).

Fifth, we control for respondents’ GPA. Individuals who have high academic performance
are likely to have a stronger belief in their ability than those who have low academic
performance (Bandura, 1993, 1997; Becker and Gable, 2009). Thus, we expect that respon-
dents having a higher GPA are less likely to choose a default investment option. Respon-
dents’ GPA, however, is excluded from the stock allocation equation because there is no
previous evidence on the relationship between GPA and the propensity to invest in stocks.
Furthermore, the exclusion of GPA from the stock allocation equation avoids any problems
for identification with the Heckman two-stage selection model.12 Thus, we assume that
respondents’ GPA has no direct effect on stock allocation.

Finally, we control for individual demographic characteristics: gender (female � 1), age,
and ethnicity (White � 1, otherwise � 0).

4. Results

4.1. Descriptive statistics

Table 3 presents descriptive statistics of the sample. The mean score of respondents’
financial literacy is 4.1, which is greater than the 3.4 that Lusardi (2011) reported for college
graduates or higher in the national survey sample of 1,488 adults in 2009. The higher score
of financial literacy may result from our sample characteristics that all respondents are
business students. For financial networks, the respondents in the sample identify, on average,
4–5 people from whom they have obtained financial or investment information. The average
financial network intensity is 0.28. This indicates that, for example, when a respondent has
five persons in his or her financial network who all are financially literate, he or she has
contacted one person at the frequency of “once a month” but the other four persons at the
frequency of “once a year or less.” For risk tolerance, our respondents have an average score
of 26.49, which is close to the average scores of 28.83 and 27.03 reported by Grable and
Lytton (2003) and Grable, Lytton, and O’Neill (2004), respectively. About 39% of our
respondents have investment experience. Female respondents consist of 41% of the sample.
Respondents’ age ranges from 20 to 26 with a mean age of 22.1. About 87% of the
respondents are White. Panel B of the table reports correlation coefficients between respon-
dents’ selection of “no default,” stock allocation, financial literacy, and financial network
variables. In particular, the variables of financial literacy, financial network size, and
financial network intensity are not significantly correlated with each other. Hence, an
inclusion of these three variables in the same regression equation would not incur significant
multi-collinearity problems among the variables.

4.2. Effects of financial literacy and financial network intensity on retirement investments

To examine the effects of financial literacy and financial networks on respondents’
investment decisions for retirement, we use the Heckman two-stage selection model de-

87Y. Chung, Y. Park / Financial Services Review 24 (2015) 77–99



scribed in Eqs. (1) and (2). Column 1 of Table 4 presents regression results of the two-stage
selection model: Column 1a presents probit regression results for decisions on the default
option (the first stage), whereas Column 1b presents ordinary least square regression results
for stock allocation (the second stage). About two-thirds of the 97 respondents select “no
default,” and they build their own retirement portfolios. Since we ask respondents to build
their retirement investment portfolios only when they answer “no default” for the default
investment option (a money market fund), the two-stage selection model would overcome a
potential selection bias problem that the error term (�i) in the stock allocation equation is
correlated with the error term (ui) in the selection equation.

First, respondents’ financial literacy is significantly positively related to a choice of “no
default” (Column 1a) and stock allocation (Column 1b). For example, if a respondent’s
financial literacy score increases from 4 to 5 (the largest score), the probability of selecting
“no default” increases by 0.189 (from 0.708 to 0.897) and stock allocation increases by 12.7
percentage points (from 48.4 to 61.0%), with the other variables held at their mean values.
The results support hypotheses H1a (a positive relationship between financial literacy and a

Table 3 Sample descriptive statistics and correlation
Panel A: Descriptive statistics

Variables Number
of obs.

Min Max Median Mean SD

Financial literacy (score) 97 2 5 4 4.09 0.84
Financial network:

Size 97 1 10 4 4.60 2.46
Intensity 97 0 0.92 0.25 0.28 0.22

Risk tolerance (score) 97 18 37 26 26.49 4.20
Investment experience (yes � 1) 97 0 1 0 0.39 0.49
Gender (female � 1) 97 0 1 0 0.41 0.49
Age 97 20 26 22 22.06 1.20
Ethnicity (White � 1) 97 0 1 1 0.87 0.34
GPA 97 1.90 4.00 3.11 3.14 0.41
Current mood state:

Positive (score) 97 17 46 37 37.24 5.22
Negative (score) 97 10 33 18 18.56 5.02

Selection of a default option (“no default” � 1) 97 0 1 1 0.66 0.48
Contribution allocation to stock funds

(for only those selecting “no default”)
64 0 1.00 0.50 0.52 0.20

Panel B: Correlations of selection of no default, stock allocation, financial literacy, and financial networks

1 2 3 4 5

1. Selection of “no default” 1.000
2. Stock allocation (for only those selecting

“no default”)
— 1.000

3. Financial literacy 0.365** 0.390** 1.000
4. Financial network size 0.096 �0.162 �0.113 1.000
5. Financial network intensity 0.128 0.399** 0.120 �0.084 1.000

**p � 0.01.

88 Y. Chung, Y. Park / Financial Services Review 24 (2015) 77–99



choice of “no default”) and H1b (a positive relationship between financial literacy and stock
allocation).

Second, financial network intensity is not significantly related to a selection of “no default”
(Column 1a), but it is significantly positively related to stock allocation (Column 1b). For
example, a respondent’s stock allocation increases by 2.7 percentage points as his or her network
intensity increases by 10%. The results do not support hypothesis H2a (a positive relationship
between financial network intensity and a choice of “no default”) but do support H2b (a positive
relationship between financial network intensity and stock allocation).

Table 4 Regression results: Default investment option and contribution allocation to stock funds

Dependent variables (1) (2)

(1a) (1b) (2a) (2b)
Selection of
no default
(first stage)

Contribution
allocation to
stock funds
(second stage)

Selection of
no default
(first stage)

Contribution
allocation to
stock funds
(second stage)

Financial literacy (score) 0.715** 0.127** 0.768** 0.111**
(0.222) (0.035) (0.235) (0.035)

Financial network size 0.104 0.006 0.109 0.003
(0.069) (0.009) (0.070) (0.008)

Financial network intensity 0.794 0.268** 0.650 0.184*
(0.751) (0.091) (0.744) (0.087)

(Financial literacy) �
(Financial network intensity)

�0.734 0.269*
(0.699) (0.107)

Risk tolerance (score) �0.033 0.017** �0.034 0.016**
(0.042) (0.005) (0.042) (0.005)

Investment experience (yes � 1) 0.938* 0.088 0.962* 0.085
(0.385) (0.049) (0.387) (0.046)

Gender (female � 1) �0.051 0.048 �0.027 0.036
(0.345) (0.041) (0.350) (0.039)

Ethnicity (White � 1) 0.011 0.162** 0.110 0.121*
(0.456) (0.061) (0.470) (0.059)

Current mood state
Positive (score) 0.011 0.012* 0.007 0.013**

(0.032) (0.004) (0.032) (0.004)
Negative (score) �0.077* �0.001 �0.079* �0.001

(0.037) (0.005) (0.037) (0.005)
GPA 1.065** 1.040*

(0.408) (0.412)
Constant �4.839* �1.281** �1.511 �0.675**

(3.389) (0.290) (1.990) (0.228)
�: Mill’s ratio 0.149 0.137
[p-value] [0.065] [0.071]
Observations 97 64 97 64
Pseudo R2 0.305 0.313
Adjusted R2 0.514 0.567

Notes: When an interaction term of Financial literacy and Network intensity is included in Column 2, the
variables of Financial literacy and Network intensity are centered with respect to the mean. This centering,
however, does not affect the coefficients and their significance. Standard errors are in parentheses.

**p � 0.01; *p � 0.05.

89Y. Chung, Y. Park / Financial Services Review 24 (2015) 77–99



Accordingly, the results reported in Columns 1a and 1b suggest that when making a
decision on whether or not to choose a default option, respondents are likely to rely on their
financial literacy but not on their financial networks. However, when constructing a retire-
ment portfolio such as determining stock allocations, respondents are likely to draw on both
their financial literacy and financial networks. Hence, financial literacy and financial network
intensity were positively associated with stock allocation.

In addition, respondents’ investment experience, current mood state, and GPA are sig-
nificantly related to their selection on the default investment option (Column 1a) or stock
allocation (Column 1b). First, when a respondent has any investment experience during the
past five years, he or she is likely to select “no default” but not likely to significantly increase
stock allocation. Second, different mood states involve different investment tasks. A negative
(or unpleasant) mood state is significantly related to a choice of the default investment option
whereas a positive (or pleasant) mood state is not. This may be because a negative mood state
inhibits people from making rational decision-making because of excessive stress (Leith and
Baumeister, 1996). In addition, a positive (pleasant) mood state is likely to increase stock
allocation in a retirement portfolio, but a negative mood state is not. Because people who
have a positive mood state tend to underestimate the possibility of loss, they are more willing
to take a risk (Au, Chan, Wang, & Vertinsky, 2003; Seo, Goldfarb, & Barret, 2010).
Respondents’ GPA is positively related to a choice of “no default,” as expected.

4.3. Interaction effects of financial literacy and financial network intensity

In addition to the main effects of financial literacy and financial network intensity on
investment decisions for retirement, we hypothesized that the two variables interact with
each other. Column 2 of Table 4 presents regression results of the Heckman two-stage
selection model that includes an interaction term of financial literacy and financial network
intensity. The interaction term is not significantly related to respondents’ selection of the
default option (Column 2a), not supporting hypothesis H3a. The interaction term, however,
is significantly positively related to stock allocation (Column 2b), supporting hypothesis
H3b. The positive sign of the interaction term indicates a complementary effect between
financial literacy and financial network intensity on stock allocation in a retirement portfolio.

To further examine the interaction effect, following Aiken and West (1991), we depict
respondents’ contribution allocation to stock funds with respect to two different levels of
financial literacy and financial network intensity. A high or low level is defined by 1 SD
above or below the mean, respectively. Fig. 1 shows that the relationship between financial
literacy and stock allocation is positive and significant in a high level of financial network
intensity (with a slope of 0.171, p-value � 0.001), but this relationship is not significant in
a low level of financial network intensity (with a slope of 0.050, p-value � 0.300). The
results suggest that the positive effects of financial literacy on stock allocation documented
in the literature (e.g., van Rooij et al., 2011; Yoong, 2010) are limited to those who have a
high level of financial network intensity (i.e., high network strength with the financially
literate).

90 Y. Chung, Y. Park / Financial Services Review 24 (2015) 77–99



4.4. Effects of financial literacy and financial network intensity on stock allocation:
Deviation from age-appropriate stock allocations

Among those who select “no default” and build their own retirement portfolio with the 10
mutual funds offered, some respondents may not construct an appropriate portfolio—for
example, too little or too much stock allocation. In this section, we attempt to see whether
respondents construct an appropriate retirement investment portfolio based on their age and
risk tolerance level, focusing on stock allocation. For doing this, we use as a benchmark the
2012 Morningstar Lifetime Allocation Indexes—U.S. Investors (Morningstar, 2012). The
Morningstar Indexes provide the information of asset allocation over lifetime for three
different risk profiles: aggressive, moderate, and conservative. We assume that the asset
allocations of the Morningstar Indexes represent age-appropriate ones for retirement.

To calculate any deviation of respondents’ stock allocation from the stock allocations of
the Morningstar Indexes, we use the following formula:

Devi � | yi � Indexk| (4)

where the variable yi indicates individual i’s contribution allocation to stock funds as a
percentage and where Indexk indicates stock allocations of the Morningstar Indexes for k �
aggressive, moderate, or conservative. A k is determined according to a respondent’s risk
tolerance level: for example, if a respondent’s risk tolerance level is low, a conservative
allocation of the Morningstar Indexes is regarded as his or her appropriate asset allocation.
A respondent’s risk tolerance is categorized into three levels based on his or her risk
tolerance score. A low level of risk tolerance is defined with scores of less than 24, a medium

Fig. 1. Interaction effects of financial literacy and network intensity on contribution allocation to stock funds.
Notes: This figure depicts the interaction effects based on the regression results in Column 2b of Table 4.

91Y. Chung, Y. Park / Financial Services Review 24 (2015) 77–99



level with scores of 24–32, and a high level with scores of more than 32. The threshold
scores of 24 and 32 are derived from the findings of Grable and Lytton (2003) and Grable
et al. (2004).13

Those who selected “no default” have constructed their retirement portfolios with a
deviation of 7.2–80.9 percentage points from the age-appropriate stock allocations. The
mean deviation is 35.8 percentage points (Panel A of Table 5). To examine how the deviation
is related to respondents’ financial literacy and financial network intensity, we estimate a
Heckman two-stage selection model that includes the deviation from the age-appropriate
stock allocations as a dependent variable in the second stage. The model also includes an
interaction term of financial literacy and financial network intensity.

Panel B of Table 5 presents the regression results for the deviation from the age-
appropriate stock allocations based on respondents’ age and risk tolerance level. Column 1b
shows that the deviation is significantly negatively related to financial literacy and financial
network intensity. In addition, the interaction term of financial literacy and financial network
intensity is significantly negatively related to the deviation. The negative sign of the
interaction term indicates a complementary effect between financial literacy and financial
network intensity on reducing the deviation from the benchmark stock allocations.

To further examine the interaction effects, we depict the deviation from the age-appro-
priate stock allocation with respect to two different levels of financial literacy and financial
network intensity. A high or low level is defined by 1 SD above or below the mean,
respectively, as in Fig. 1. Fig. 2 shows that the effect of financial literacy on the deviation
is significant in the high financial network intensity (with a slope of �0.156, p-value
� 0.001), but is not significant in the low financial network intensity (with a slope of �0.056,
p-value � 0.212). The results indicate that a reduction of the deviation from the age-
appropriate stock allocation by enhancing financial literacy would be effective only among
those who have a high level of financial network intensity (i.e., strong relationships with the
financially literate).

5. Conclusion

This article examines the role of financial literacy and financial network intensity in
retirement investment decisions of Generation Yers. With a complete sample of 97 senior
business college students, aged 20–26 in 2012, we find first that when a money market fund
is designated as a default investment option, respondents who have high financial literacy are
more likely to select “no default,” but their financial networks are not significantly related to
their choice. Second, respondents who have high levels of financial literacy or financial
network intensity are likely to allocate more contributions to stock. Third, financial literacy
and financial network intensity significantly interact on stock allocation in a retirement
portfolio. Specifically, the positive effect of financial literacy on stock allocation is signif-
icant in the high financial network intensity, but the financial literacy effect is not significant
in the low financial network intensity. Last, when respondents construct their retirement
portfolios, stock allocations are largely deviated from the age-appropriate stock allocations
given their age and risk tolerance level. The deviations, however, are significantly reduced

92 Y. Chung, Y. Park / Financial Services Review 24 (2015) 77–99



when financially knowledgeable respondents have strong relationships with the financially
literate.

Our findings contribute to an understanding of how financial literacy and financial
networks are jointly associated with individual investors’ decision-making. Although most
previous studies argue that financial literacy is a key determinant of individual investors’
decision-making, our study shows that both financial literacy and financial network intensity

Table 5 Contribution allocation to stock funds: Deviation from age-appropriate allocations
Panel A: Descriptive statistics

obs. Min Max Median Mean SD

Deviation from age-appropriate stock allocations 64 0.072 0.809 0.383 0.358 0.174

Panel B: Regression results

Dependent variables (1-a) (1-b)

Selection of
no default
(first stage)

Deviation of
stock allocation
(second stage)

Financial literacy (score) 0.768** �0.106**
(0.235) (0.032)

Financial network size 0.109 �0.001
(0.070) (0.008)

Financial network intensity 0.650 �0.169*
(0.744) (0.080)

(Financial literacy) � (Financial network intensity) �0.734 �0.223*
(0.699) (0.097)

Risk tolerance (score) �0.034 �0.009
(0.042) (0.005)

Investment experience (yes � 1) 0.962* �0.077
(0.387) (0.042)

Gender (female � 1) �0.027 �0.037
(0.350) (0.036)

Ethnicity (White � 1) 0.110 �0.103
(0.470) (0.054)

Current mood state:
Positive (score) 0.007 �0.013**

(0.032) (0.004)
Negative (score) �0.079* �0.001

(0.037) (0.005)
GPA 1.040*

(0.412)
Constant �1.511 1.345**

(1.990) (0.209)
�: Mill’s ratio �0.135
[p-value] [0.051]
Observations 97 64
Pseudo R2 0.313
Adjusted R2 0.513

Notes: The variables of financial literacy and network intensity are centered with respect to the mean. This
centering, however, does not affect the coefficients and their significance. Standard errors are in parentheses.

**p � 0.01; *p � 0.05.

93Y. Chung, Y. Park / Financial Services Review 24 (2015) 77–99



are critical factors influencing individuals’ investment decision-making. In particular, the
significant interaction between financial literacy and financial network intensity suggests that
financial literacy and financial networks are complementary to each other. This finding
implies that an individual’s retirement investment decisions can be significantly enhanced
when his or her financial knowledge is synthesized with the knowledge acquired through
social networks with the financially literate. This result also provides an insight into
employer-sponsored financial education. For retirement investments, individual investors
may not always use the information provided by employers, but rather rely on their social
networks (e.g., Duflo & Saez, 2002, 2003). Thus, for those who are likely to lack social
networks with the financially literate, employers need to provide more opportunities to build
networks for financial advice with financial planners or advisors because financial planners
or advisors may be able to be substituted for their financial networks.

Although this study contributes to the literature that pertains to financial literacy, social
interactions, and investment behavior in retirement plans, it has several limitations. First, this
study does not use actual contribution allocations of defined contribution plan participants.
Thus, the results from our sample may not represent actual investing behavior of defined
contribution plan participants. Second, because of the small sample size, the results in the
study may not represent the investing behavior of Generation Yers as a whole. Third, because
the sample consists of college students who tend to have a higher level of risk tolerance (e.g.,
Gilliam, Chatterjee, & Grable, 2010) and have less investing experience than the general
investing population, the results may be biased towards allocating more contributions to
risky assets. Despite these limitations, our study using an experimental survey design (i.e.,

Fig. 2. Interaction effects of financial literacy and network intensity on the deviation from age-appropriate stock
allocations. Notes: This figure depicts the interaction effects based on the regression results in Column 1b, Panel
B of Table 5.

94 Y. Chung, Y. Park / Financial Services Review 24 (2015) 77–99



providing respondents a scenario of retirement investments and having them make invest-
ment decisions) provides an implication for Generation Yers’ retirement investments: finan-
cial literacy combined with financial networks could increase the effectiveness of any efforts
for enhancing their investment decisions. This could also be applied to improving long-term
financial planning such as budgeting and saving. Therefore, future research is called for to
collect field data from those who engage in the first-time retirement investment or long-term
financial planning and investigate how their financial literacy and financial networks jointly
influence their decision-making quality.

Notes

1 Several recent articles provide excellent literature reviews on the effects of financial
literacy and financial education on financial outcomes (e. g., Collins & O’Rourke,
2010; Fernandes, Lynch, & Netemeyer, 2014; Hastings, Madrian, & Skimmyhorn,
2013).

2 Prior studies on social interactions focused on social activities, such as visiting
neighbors and attending church (Hong et al., 2004), asking for advice about products
and brands (Brown et al., 2008), and participating in a sport or social club, or a
political or community-related organization (Christelis, Jappelli, & Padula, 2010).

3 In this article, respondents’ financial literacy is measured with their financial knowl-
edge, as in most prior studies on financial literacy (e. g., Alhenawi & Elkhal, 2013;
Collins, 2012; Lusardi & Mitchell, 2007a; Robb, Babiarz, & Woodyard, 2012; van
Rooij et al., 2011; Yoong, 2010), while respondents’ financial network intensity is
measured with their network strength with financially literate ones. We will discuss
financial literacy and financial network intensity in more detail in Section 3.

4 The complementary effects of financial literacy and financial network intensity on
contribution allocation to stock funds are similar to the finding of Collins (2012). He
finds that individuals with a higher level of financial literacy are more likely to receive
financial advice.

5 Several studies on stock-market participation show that information and transaction
costs deter people from participating in the stock market (e.g., Haliassos & Bertaut,
1995; Vissing-Jorgensen, 2003).

6 For example, Vissing-Jorgensen (2003) mentions that once people understand diver-
sification, they can apply their insights to a larger portfolio without cost.

7 Van Rooij et al. (2011) use the information of respondents’ economics education in
school as an instrument for financial literacy and find a positive relationship between
financial literacy and stock-market participation.

8 Guiso et al. (2008) explore a relationship between trust and stock market participation.
Using Dutch and Italian survey data, they find that less trusting individuals are less
likely to be stockholders.

9 Since this survey was conducted in a context of a 401(k) plan, the survey participants
were informed before the survey that they could rebalance their retirement portfolio
later at no cost.

95Y. Chung, Y. Park / Financial Services Review 24 (2015) 77–99



10 For the exact wording of the five questions, see the survey questions posted to the
website of the National Financial Capability Study, http://www.usfinancialcapa
bility.org/survey_data.html.

11 We use the items of Grable and Lytton (1999), instead of questions about hypothet-
ical gambles over lifetime income in the Health and Retirement Study (HRS),
because most of our survey participants may not be familiar with the questions used
in the HRS because of their little or limited full-time work experience. Our respon-
dents’ median (mean) full-time work experience is only 0.5 years (1.8 years).

12 We include GPA in the selection equation but not in the stock allocation equation.
The variable GPA works as an exclusion restriction. The Heckman two-stage selec-
tion model should have at least one exclusion restriction; otherwise, the second stage
of Heckman two-stage model is likely to suffer from a collinearity problem, which
provides imprecise estimates as a result (Wooldridge, 2002).

13 Grable and Lytton (2003) report an average score of 28.83 and a SD of 4.49 with a
sample of 303 respondents, while Grable et al. (2004) report an average score of
27.03 and a SD of 5.18 with a sample of 421 respondents. As a range of a medium
level of risk tolerance, we use the common interval of �one SD from the two studies,
which is a range of 24–32.

Acknowledgments

The authors would like to thank Michele O’Neill and anonymous referees for their helpful
comments.

References

Agnew, J. R., & Szykman, L. R. (2005). Asset allocation and information overload: The influence of information
display, asset choice, and investor experience. Journal of Behavioral Finance, 6, 57–70.

Aiken, L. S., & West, S. G. (1991). Multiple regression: Testing and interpreting interactions. Newbury Park,
CA: Sage.

Alhenawi, Y., & Elkhal. (2013). Financial literacy of U.S. households: Knowledge vs. long-term financial
planning. Financial Services Review, 22, 211–244.

Au, K., Chan, F., Wang, D., & Vertinsky, I. (2003). Mood in foreign exchange trading: Cognitive processes and
performance. Organizational Behavior and Human Decision Processes, 91, 322–338.

Balkundi, P., & Harrison, D. A. (2006). Ties, leaders, and time in teams: Strong inference about network
structure’s effects on team viability and performance. Academy of Management Journal, 49, 49–68.

Bandura, A. (1993). Perceived self-efficacy in cognitive development and functioning. Educational Psychologist,
28, 117–148.

Bandura, A. (1997). Self-efficacy: The exercise of control. New York, NY: W. Freeman.
Banerjee, A. (1992). A simple model of herd behavior. Quarterly Journal of Economics, 107, 797–817.
Banerjee, S. (2011). How do financial literacy and financial behavior vary by state? Employee Benefit Research

Institute Note, 32, 9–13.
Becker, S. P., & Gable, R. K. (2009). The relationship of self-efficacy with GPA, attendance, and college student

retention. NERA Conference Proceedings, Paper 26.

96 Y. Chung, Y. Park / Financial Services Review 24 (2015) 77–99



Bettman, J. R., & Park, C. W. (1980). Effects of prior knowledge and experience and phase of the choice process
on consumer decision processes: A protocol analysis. Journal of Consumer Research, 7, 234–248.

Bikchandani, S., Hirshleifer, D., & Welch, I. (1992). A theory of fads, fashion, custom and cultural change as
informational cascades. Journal of Political Economy, 100, 992–1026.

Brown, J. R., Ivković, Z., Smith, P. A., & Weisbenner, S. J. (2008). Neighbors matter: Causal community effects
and stock market participation. Journal of Finance, 63, 1509–1531.

Campbell, J. (2006). Household finance. Journal of Finance, 61, 1553–1604.
Cascioaro, T., & Lobo, M. S. (2008). When competence is irrelevant: The role in interpersonal affect in

task-related ties. Administrative Science Quarterly, 53, 655–684.
Choi, J. J., Laibson, D., Madrian, B. C., & Metrick, A. (2004). For better or for worse: Default effects and 401(k)

savings behavior. In: D. A. Wise (Ed.), Perspectives on the Economics of Aging (pp. 81–121). Chicago, IL:
University of Chicago Press.

Choi, J. J., Laibson, D., Madrian, B. C., & Metrick, A. (2006). Saving for retirement on the path of least
resistance. In: E. J. McCaffrey & J. Slemrod (Eds.), Behavioral Public Finance: Toward a New Agenda (pp.
304–351). New York, NY: Russell Sage Foundation.

Christelis, D., Jappelli, T., & Padula, M. (2010). Cognitive abilities and portfolio choice. European Economic
Review, 54, 18–38.

Chung, Y., & Jackson, S. E. (2011). Coworker trust and knowledge creation: A multilevel analysis. Journal of
Trust Research, 1, 65–83.

Chung, Y., & Jackson, S. E. (2013). The internal and external networks of knowledge-intensive teams: The role
of task routineness. Journal of Management, 39, 442–468. 4

Collins, J. (2012). Financial advice: A substitute for financial literacy? Financial Services Review, 21, 307–322.
Delavande, A., Rohwedder, S., & Willis, R. (2008). Preparation for Retirement, Financial Literacy and

Cognitive Resources. Michigan Retirement Research Center Working Paper, WP2008-190.
Duflo, E., & Saez, E. (2002). Participation and investment decisions in a retirement plan: The influence of

colleagues’ choices. Journal of Public Economics, 85, 121–148.
Duflo, E., & Saez, E. (2003). The role of information and social interactions in retirement plan decisions:

Evidence from a randomized experiment. Quarterly Journal of Economics, 118, 815–842.
Ellison, G., & Fudenberg, D. (1993). Rules of thumb for social learning. Journal of Political Economy, 101,

93–126.
Ellison, G., & Fudenberg, D. (1995). Word of mouth communication and social learning. Quarterly Journal of

Economics, 110, 93–125.
Fernandes, D., Lynch, J. G., & Netemeyer, R. G. (2014). Financial literacy, financial education, and downstream

financial behaviors. Management Science, 60, 1861–1883.
Forgas, J. P. (1995). Mood and judgment: The affect infusion model (AIM). Psychological Bulletin, 117, 39–66.
Gargiulo, M., Ertug, G., & Galunic, C. (2009). The two faces of control: Network closure and individual

performance among knowledge workers. Administrative Science Quarterly, 54, 299–333.
Gibson, R., Michayluk, D., & Venter, G. (2013). Financial risk tolerance: An analysis of unexplored factors.

Financial Services Review, 22, 23–50.
Gilliam, J., Chatterjee, S., & Grable, J. (2010). Measuring the perception of financial risk tolerance: A tale of two

measures. Journal of Financial Counseling and Planning, 21, 30–43.
Grable, J. E., & Lytton, R. H. (1999). Financial risk tolerance revisited: The development of a risk assessment

instrument. Financial Services Review, 8, 163–181.
Grable, J. E., & Lytton, R. H. (2003). The development of a risk assessment instrument: A follow-up study.

Financial Services Review, 12, 257–274.
Grable, J., Lytton, R., & O’Neill, B. (2004). Projection bias and financial risk tolerance. Journal of Behavioral

Finance, 5, 142–147.
Guiso, L., Sapienza, P., & Zingales, L. (2008). Trusting the stock market. Journal of Finance, 63, 2557–2600.
Haliassos, M., & Bertaut, C. (1995). Why do so few hold stocks? Economic Journal, 105, 1110–1129.
Hansen, M., Podolny, J. M., & Pfeffer, J. (2001). So many ties, so little time: A task contingency perspective on

97Y. Chung, Y. Park / Financial Services Review 24 (2015) 77–99



the value of social capital in organisations. In: S. M. Gabbay & R. Th. A. J. Leenders (Eds.), Research in the
Sociology of Organizations (Vol. 18, pp. 21–57). London: JAI Press.

Hansen, M. T., Mors, M. L., & Lovas, B. (2005). Knowledge sharing in organizations: Multiple networks,
multiple phases. Academy of Management Journal, 48, 776–793.

Hastings, J. S., Madrian, B. C., & Skimmyhorn, W. L. (2013). Financial literacy, financial education, and
economic outcomes. Annual Review of Economics, 5, 347–373.

Hilgert, M. A., Hogarth, J. M., & Beverly, S. G. (2003). Household financial management: The connection
between knowledge and behavior. Federal Reserve Bulletin, 7, 309–322.

Hong, H., Kubik, J. D., & Stein, J. C. (2004). Social interaction and stock-market participation. Journal of
Finance, 59, 137–163.

Hung, A., Parker, A. M., & Yoong, J. (2009). Defining and Measuring Financial Literacy. RAND Labor and
Population Working Paper, No. 708.

Isen, A. M. (2000). Positive affect and decision making. In: J. M. Haviland (Ed.), Handbook of Emotions (pp.
261–277). London: Guilford Press.

Ivković, Z., & Weisbenner, S. J. (2007). Information diffusion effects in individual investors common stock
purchases: Covet thy neighbors investment choices. Review of Financial Studies, 20, 1327–1357.

Kaustia, M., & Knüpfer, S. (2012). Peer performance and stock market entry. Journal of Financial Economics,
104, 321–338.

Kiviat, R., & Morduch, J. (2012). From Financial Literacy to Financial Action. Working paper, McGraw-Hill
Research Foundation.

Krackhardt, D. (1992). The strength of strong ties: The importance of philos in organizations. In: N. Nohria &
R. Eccles (Eds.), Networks and Organizations: Structure, Form, and Action (pp. 216–239). Boston, MA:
Harvard Business School Press.

Leith, K. P., & Baumeister, R. F. (1996). Why do bad moods increase self-defeating behavior? Emotion,
risk-taking, and self-regulation. Journal of Personality and Social Psychology, 71, 1250–1267.

Loewenstein, G., Weber, E. U., Hsee, C. K., & Welch, N. (2001). Risk as feelings. Psychological Bulletin, 127,
267–286.

Lu, T. (2011). Social Interaction Effects and Individual Portfolio Choice: Evidence From 401(k) Pension Plan
Investors. Pension Research Council Working Paper, PRC WP2011-19.

Lusardi, A. (2011). Americans’ Financial Capability. Pension Research Council Working Paper, PRC WP2011-
02.

Lusardi, A., & Mitchell, O. S. (2007a). Financial Literacy and Retirement Planning: New Evidence From the
Rand American Life Panel. Michigan Retirement Research Center Working Paper, WP2007-157.

Lusardi, A., & Mitchell, O. S. (2007b). Baby boomer retirement security: The roles of planning, financial literacy,
and housing wealth. Journal of Monetary Economics, 54, 205–224.

Lusardi, A., & Tufano, P. (2009). Debt Literacy, Financial Experiences, and Over-indebtedness. Cambridge,
MA: National Bureau of Economic Research, NBER Working Paper 14808.

Madrian, B. C., & Shea, D. F. (2001). The power of suggestion: Inertia in 401(k) participation and saving
behavior. Quarterly Journal of Economics, 116, 1149–1187.

McFadyen, M. A., & Cannella, A. A. (2004). Social capital and knowledge creation: Diminishing returns of the
number and strength of exchange relationships. Academy of Management Journal, 47, 735–746.

Moore, D. (2003). Survey of Financial Literacy in Washington State: Knowledge, Behavior, Attitudes, and
Experiences, Pullman, WA: Washington State University Social and Economic Sciences Research Center,
Technical Report 03-39.

Morningstar. (2012). Morningstar Lifetime Allocation Indexes—U. S. Investors. Chicago, IL: Morningstar, Inc.
Morrison, E. W. (2002). Newcomers’ relationship: The role of social network ties during socialization. Academy

of Management Journal, 45, 1149–1160.
Nahapiet, J., & Ghoshal, S. (1998). Social capital, intellectual capital, and the organizational advantage. Academy

of Management Review, 23, 242–266.
Perry, V. G., & Morris, M. D. (2005). Who is in control? The role of self-perception, knowledge, and income in

explaining consumer financial behavior. Journal of Consumer Affairs, 39, 299–313.

98 Y. Chung, Y. Park / Financial Services Review 24 (2015) 77–99



Reinholt, M., Pedersen, T., & Foss, N. J. (2011). Why a central network position isn’t enough: The role of
motivation and ability for knowledge sharing in employee networks. Academy of Management Journal, 54,
1277–1297.

Robb, C. A., Babiarz, P., & Woodyard, A. (2012). The demand for financial professionals’ advice: The role of
financial knowledge, satisfaction, and confidence. Financial Services Review, 21, 291–305.

Social Security and Medicare Boards of Trustees. (2012). The 2012 Annual Report of the Trustees of the Federal
Old-Age and Survivor’s Insurance and Disability Insurance Trust Funds. Baltimore, MD: Social Security
Administration.

Seo, M., Goldfrab, B., & Barrett, L. F. (2010). Affect and the framing effect within individuals over time: Risk
taking in a dynamic investment simulation. Academy of Management Journal, 53, 411–431.

van Rooij, Lusardi, M., A., & Alessie, R. (2011). Financial literacy and stock market participation. Journal of
Financial Economics, 101, 449–472.

van Rooij, Lusardi, M., A., & Alessie, R. (2012). Financial literacy, retirement planning and household wealth.
Economic Journal, 122, 449–478.

Vissing-Jørgensen, A. (2003). Perspectives on behavioral finance: Does “irrationality” disappear with wealth?
Evidence from expectations and actions. Cambridge, MA: National Bureau of Economic Research Macro-
economics Annual.

Wang, S., & Noe, R. A. (2010). Knowledge sharing: A review and directions for future research. Human
Resource Management Review, 20, 115–131.

Watson, D., Clark, L. A., & Tellegen, A. (1988). Development and validation of brief measures of positive and
negative affect: The PANAS scales. Journal of Personality and Social Psychology, 54, 1063–1070.

Willis, L. (2011). The financial education fallacy. American Economic Review, 101, 429–434.
Wooldridge, J. M. (2002). Econometric Analysis of Cross Section and Panel Data. Cambridge, MA: MIT Press.
Yli-Renko, H., Autio, E., & Sapienza, H. J. (2001). Social capital, knowledge acquisitions, and knowledge

exploitation in young technology-based firms. Strategic Management Journal, 22, 587–613.
Yoong, J. (2010). Financial Illiteracy and Stock Market Participation: Evidence From the RAND American Life

Panel. Pension Research Council Working Paper, PRC WP2010-29.

99Y. Chung, Y. Park / Financial Services Review 24 (2015) 77–99