1.1 Research Background
In the U.S. retirement system, defined contribution plans play an important role in
helping Americans prepare for their retirement. The 401(k) plan, one major type of the
defined contribution plans, is typically sponsored by employers. Employees can select
different investment options, usually mutual funds, provided by sponsors. Other
investment options that might be available in 401(k) plans include: company stock,
guaranteed investment contracts (GICS), and ETFs. According to the 2011 Investment
Company Fact Book, 401(k) plans hold $3.1 trillion in assets at the year-end 2010,
with 59 percent of 401(k) assets invested in mutual funds. It indicates that mutual
funds have been an important long-term investment vehicle for 401(k) plan
participants. Therefore, the selection of mutual funds into 401(k) plans is an important
issue for employers (sponsors) and participants.
Liquidity advantage of mutual funds, arguably the primary service of open-end
mutual funds, provides a great deal of liquidity to investors. Investors can redeem
their shares at the fund’s daily-close net asset value on any given day. This
characteristic of liquidity provided by open-end mutual funds, however, may have
negative impact on future fund performance. Following substantial outflows, mutual
2
uninformed, liquidity-motivated trading, which results in an adverse effect on
subsequent fund return. Based on prior literature (e.g., Edelen, 1999), the costs
involved in such trade including direct costs（e.g., commissions）and indirect costs in
the form of a negative relation between a fund’s abnormal return and investor flows.
Nevertheless, such costs due to large redemptions are not reflected on the day of
redemption but borne by the remaining investors who still stay in the funds. Such
adverse effect is referring to investor externality.
1.2 Objectives
In the current study, we are interested in whether 401(k) participants suffer from
the negative impact of investor externality in terms of lower buy-and-hold return
when they invest in the mutual funds included in 401(k) plans (referred to 401(k)
mutual fund hereafter). In particular, considering that the 401(k) participants are
investors with long-term horizons, when the funds in 401(k) plans encounter flow
shocks, we hypothesize that participants in 401(k) retail-oriented funds, compared to
institutional-oriented funds, would suffer higher negative investor externality in terms
of lower future buy-and-hold return. The rationale is due to the different redemption
pattern between the retail investors and institutional investors documented by Chen,
Goldstein, and Jiang (2010). They present the concept of “strategic
complementarities”- the expectation that other investors will withdraw their money
reduces the expected return from staying in the fund and increases the incentive for
each individual investor to withdraw as well. Retail investors are more likely than
institutional investors to subject to such strategic complementarities since retail
investors have only small shares and therefore their redemption decisions are affected
by what they believe other investors will do. To illustrate, when confronting large
outflows, if retail investors do not withdraw, they would bear the strategic risk due to
the externalities from other investors’ redemptions. As each investor has the
expectation of other investors’ withdraw, everyone redeems their shares as well and
finally amplifies the damage of the redemptions. Such self-fulfilling redemption
behaviors documented by Chen, Goldstein, and Jiang (2010) therefore could hurt the
401(k) participants with long-term horizons who stay in the fund. As a result, we
suggest that when encountering large outflows, retail investors in retail funds would
rush to redeem their shares, and therefore impose significant costs on those 401(k)
investors who stay in the retail-oriented funds. On the other hand, in
institutional-oriented funds, the investors have relatively stable, long-term
investments and hold large shares, they are thus less likely to exhibit self-fulfilling
redemption behaviors when facing outflows. Thus we suggest that for 401(k)
participants who stay in the fund, they may also less likely to bear the negative impact
4
future buy-and-hold returns compared to the funds with low past investor externality
and thus it’s unfavorable for 401(k) participants who stay in the fund. The logic is that,
some specific funds, for example, small-cap, mid-cap or single-country international
stock funds which are documented by Chen, Goldstein, and Jiang (2010), are proved
to suffer the impact of strategic complementarities due to the illiquid assets held by
these funds. Therefore, as outflow shocks happen, the redemptions caused by these
outflows may impose significant costs on the funds which subsequently damage the
fund performance. Such negative investor externality is then harmful for the 401(k)
participants who put their money in the 401(k) accounts.
Using data from the U.S. 401(k) plans, we find that retail-oriented funds included
in 401(k) plans , compared to institutional-oriented funds included in 401(k) plans,
have higher investor externality and have lower buy-and-hold abnormal return. The
1 Large investors could still redeem more shares for information reasons. The idea we want to emphasize is that they behave differently from retail investors in respond to outflows.
buy-and-hold abnormal return for retail-oriented funds is statistically significant lower
than the institutional-oriented funds about 7%. Second, compared to the funds with
low past investor externality, the funds with high past investor externality in the 401(k)
plans exhibit higher investor externality in terms of lower buy-and-hold abnormal
return.
1.3 Organization of the thesis
The organization of the paper is as follows. In Section 2 develops the preceding
arguments more fully and presents our main hypotheses. In section 3, we describe the
data used for our empirical study and the measure we used for evaluating the investor
externality. In section 4, we provide the empirical results. Section 5 concludes.
CHAPTER 2- HYPOTHESES DEVELOPEMENT