The evidence on target-date mutual funds

Sandeep Singha,*
aDepartment of Accounting, Economics and Finance, School of Business Administration and Economics,

College at Brockport, SUNY, Brockport, NY 14420, USA

Abstract

This paper assimilates the knowledge and evidence on target-date mutual funds (TDFs). It begins
with a discussion of the environment that contributes to the tremendous growth of TDFs. Next, a
survey of the theory and recommendations on glide paths indicates a trend towards focusing on
meeting retirement liabilities, rather than optimizing asset only portfolios. A review of performance
evaluation metrics for TDFs shows that none of the available indexes possesses all seven character-
istics of an ideal benchmark. Plan sponsors can provide better outcomes by offering multiple risk
profile TDFs while researchers can focus on improving glide path and benchmark design. © 2016
Academy of Financial Services. All rights reserved.

JEL classification: J26; C00; D14; G11

Keywords: Target-date funds; Life cycle funds; Retirement; Asset allocation

1. Introduction

The Department of Labor’s safe harbor provision of 2007 has provided a significant boost
to the popularity of target date mutual funds (TDFs). These funds are now included as a
default investment option in defined contribution plans along with managed accounts and
balanced funds. According to Department of Labor, the Qualified Default Investment
Alternative (QDIA) provides a plan sponsor “safe harbor relief from fiduciary liability for
investment outcomes EBSA (2008).” A consequence of the Pension Protection Act of 2006,
employers are increasingly adopting the automatic enrollment option in 401(K) plans, which
further supplements the assets under management (AUM) base of TDFs.

* Corresponding author. Tel.: �1-585-395-5519; fax: �1-585-395-2542.
E-mail address: ssingh@brockport.edu

Financial Services Review 25 (2016) 235–262

1057-0810/16/$ – see front matter © 2016 Academy of Financial Services. All rights reserved.



Investors held about $1 trillion in target date and lifestyle funds at the end of 2014,
compared to a total of $2.75 billion at the end of 1995 (ICI Handbook, 2015). This calculates
to a compounded annual growth rate (CAGR) of about 35% over the 10-year period, an
impressive flow of investor money to these funds.1

A TDF is managed from its purchase date through an expected retirement year, and in
some cases, the fund provides an additional “during retirement” investment option. On the
other hand, a lifestyle fund is directed to a broad age group and tailored to its purported risk
tolerance. For example, a lifestyle fund focused on younger investors in their 30s will usually
have a high equity exposure, while a lifestyle fund aimed at retirees will likely have a heavy
emphasis on fixed income securities. Assuming that one’s risk tolerance decreases with age,
an investor will move from one lifestyle fund to the next one catering to their “lifestyle” as
they age, until they are in retirement. In the case of a TDF, this asset allocation is
automatically shifted for the investor. A good way to understand the difference between
target date and lifestyle funds is to think of lifestyle funds as building blocks of a TDF. So,
which of these two fund types should investors prefer? Chang et al. (2014) use a utility
maximization framework and use bootstrap simulations to compare welfare benefits of both
types of funds. The primary focus of their research is to measure utility derived from fixed
and decreasing equity allocation for an individual investor over time. They find that a
decreasing equity allocation provides better welfare benefits than a static one. This implies
that TDFs are superior investment vehicles from a utility maximization perspective. The
authors caution that there is no one-size-fits all TDF, the investor should select such funds
based on their risk tolerance. For the purposes of this paper, the term “target date fund” is
used to designate a broad class of open-end mutual funds and exchange traded funds (ETFs)
that include both target date and lifestyle funds.

The appeal of a target-date mutual fund lies in the convenience it provides to the investor.
She does not have to monitor, and periodically alter the asset allocation because of passage
of time. In most cases, the fund mandate automatically provides for that. Based on the chosen
retirement age, the fund manager allocates the funds to a predetermined allocation schedule;
say 85% stocks and 15% fixed income initially when the investor is young. This allocation
may eventually reverse to 15% stocks and 85% fixed income closer to the “target date.” This
change in asset allocation that occurs over many years is commonly referred to as the fund’s
“glide path.”

The Vanguard Target Retirement Fund 2055 (Ticker: VFFVX), is provided as an example.
The 2055 in the fund name indicates the anticipated retirement year for the investor. In 2015,
this fund is recommended for a 30-year old anticipating retirement at age 70 or a 25-year old
anticipating retirement at age 65. Based on the fund’s glide path, the initial asset allocation
is quite aggressive, with up to 90% in equity securities. According to Vanguard, this becomes
“more conservative over time, meaning that the percentage of assets allocated to stocks will
decrease while the percentage of assets allocated to bonds and other fixed income invest-
ments will increase.” After the year 2055, the fund will mimic the allocation of Vanguard Target
Retirement Income Fund (Ticker: VTINX), another fund in the Vanguard funds family. For a
comparison of the two funds, the reader is referred to Table 1 (Vanguard Group [2015]).

The glide path of the Vanguard Target Retirement Funds is shown in Fig. 1. As expected,
the glide path of the fund becomes more conservative as the target date approaches. Based

236 S. Singh / Financial Services Review 25 (2016) 235–262



on the fund’s prospectus the manager of the fund has significant discretion in the asset
allocation decision. For example, while the allocation to equities is required to diminish from
90% to 30% by the retirement date, the manager has discretion of plus or minus seven years
around the target date to accomplish that. Such discretion may not be the case for all target
date mutual funds. Many have a more rigid timetable for asset allocation change.

There are multiple approaches to designing TDFs with each fund family emphasizing their
approach as superior, perhaps in the pursuit of market share. Some families use index funds
in the lineup of TDFs while others rely on actively managed ones. Some have rigid or
semi-rigid asset allocations while others provide their managers with broader latitude in such
decisions. The wide range of design approaches can be illustrated by comparing the Van-

Table 1 Comparative information on Vanguard Target Retirement fund 2055 and target retirement income
funda

Composition of funds Vanguard Target
Retirement 2055 fund

Vanguard Target
Retirement Income fund

Ticker VVFVX VTINX
Assets under management $1.67 billion $11.21 billion
Target equity 90% 30%
Target fixed income and cash 10% 70%
Acquired fund fees and expenses 0.18% 0.16%
Asset allocations as of 9/30/2014
Total Stock Market Index 63.0% 21.1%
Total International Stock Market Index 27.0% 8.8%
Total Bond Market II Index 8.0% 39.3%
Total International Bond Index fund 2.00% 14.0%
Short-term inflation-protected securities 0.0% 16.8%

Note: aReported for September 30, 2014.

Fig. 1. An illustrative glide path for Vanguard target date fund.

237S. Singh / Financial Services Review 25 (2016) 235–262



guard funds to PIMCO funds. PIMCO TDFs utilize a liability driven investing (LDI)
approach to asset allocation changes. The design is motivated by the goal of meeting
inflation-adjusted withdrawals (liabilities) during retirement (a more in-depth discussion on
Liability Driven Investing [LDI] is provided in a later section of this paper). Fig. 2 depicts
the glide path of PIMCO TDFs. The wide range of approaches to TDF design is starkly
contrasted by Figs. 1 and 2. The key difference that can be observed is the exposure to
inflation insensitive assets like TIPS, commodities and, real estate during retirement in the
two glide paths.

To the best of my knowledge, this is the first paper that provides a comprehensive review
and evaluation of the body of knowledge on TDFs. The focus of this paper is four fold. In
the first section, a discussion of the current state of the target date mutual fund market is
provided. The discussion includes analysis of factors that have aided in the growth of this
segment of the investment companies market. In the second section, a discussion of the
theoretical basis and a summary of varied suggestions for designing glide paths are pre-
sented. In the third section performance and benchmarking of these funds is discussed. In the
fourth, recommendations and suggestions for plan sponsors, financial advisors and research-
ers are made. Concluding remarks are provided in the final section.

2. The TDF market

Over the 10-year period 2005–2014, assets under management in TDFs have grown about
five-fold. According to the Investment Company Institute (ICI), investors had a choice of 797
TDFs for a total of 3,036 mutual fund share classes at the end of 2014 (ICI [2015]). Bauer,

Fig. 2. Glide path of PIMCO target-date funds.

238 S. Singh / Financial Services Review 25 (2016) 235–262



Phillips, and White (2009) predict that 80% of all new defined contribution plan investments
will go to TDFs and the target-date market will be $2.3 trillion by end of 2018. There are
three broad factors playing a role in the remarkable growth of this market namely, the
regulatory environment, demographics, and investor affinity for passive investments.

Before TDFs became popular, Agnew et al. (2003) studied 7,000 accounts of a large
defined contribution plan over the period 1994 to1998. They find significant participant
indifference towards the management of their retirement assets. Most asset allocations are at
extremes, either 100% or 0% equities. They also find “very limited portfolio reshuffling”
implying that rebalancing of portfolios is rare. More alarmingly, in a study of over a million
retirement accounts from about 1,000 pension plans, Tang and Mitchell (2010) find that most
employers provide an efficient menu of investment choices in pension plans. However,
participants undo this by making inefficient portfolio selection decisions. This costs the
participants a reduction of about one-fifth in potential retirement wealth. In addition
to participant education, the natural corollary of these findings would be to have investment
vehicles in retirement plans that do not require significant participant intervention over time,
that is “do no harm.” Each of the preceding studies shows that fund owners are not likely to
manage their retirement funds to their best advantage. This provides a compelling argument
in favor of TDFs. Funds that absolve the portfolio owner of monitoring and periodic
adjustments to reflect altered risk tolerance.

The most significant boost to TDFs comes from two regulatory developments: (1) the
automatic enrollment (opt-out option) feature under the Pension Protection Act of 2006 and
(2) the Department of Labor directive of 2007 that provides safe harbor relief to plan
sponsors if a TDF is offered as a Qualified Default Investment Option (QDIA). The other two
QDIA options that emerged from the Pension Protection Act of 2006 (PPA) are managed
accounts and balanced funds EBSA (2013). This regulatory protection has greatly increased
the popularity of TDFs in qualified defined contribution retirement plans.

According to Landsberg (2007), PPA provides changes in five broad areas: defined benefit
plan funding, hybrid qualified plans, defined contribution plans, executive compensation, and
plan distributions. Two broad set of key changes promote the growth of assets invested in
TDFs. The first relates to enrolment procedures and regulatory relief. This includes permis-
sion to employers to offer automatic enrollment in 401(K) plans and safe harbor provisions
for default options that include TDFs in automatic enrollment plans. A default investment
option becomes the deemed choice of a participant when no selection of investment
alternatives is specifically made. The second contributing factor makes permanent the
provisions of Economic Growth and Tax Relief Reconciliation Act (EGTRRA) of 2001 with
respect to increased contributions, inflation indexing and “catch up” provisions for contri-
butions to various Individual Retirement Arrangements (IRAs) and 401(K) plans.

Balduzzi and Ruetter (2015) report increased heterogeneity in risk taking and associated
returns of target date mutual funds since the passing of PPA of 2006. They attribute this
divergence in performance to risk taking by fund families with low market shares attempting
to differentiate themselves to attract plan sponsors. Consequently, participants in different funds
of the same vintage year might experience returns significantly different from the average.

In a study of its participants, TIAA-CREF researchers Richardson and Bissette (2014) find
that over the period 2005–2011 TDFs garnered a significant proportion (22%) of all

239S. Singh / Financial Services Review 25 (2016) 235–262



contributions. These funds are the preferred choice of younger participants with lower
account balances and are utilized as a fund of funds strategy, meaning that the target date
fund is the primary investment of the participant. A similar trend is reported for the broader
401(K) market by ICI. At the end of 2013, 70% of 401(K) plans offered TDFs and the
proportion of recently hired participants in their 20s and 30s holding TDFs is 51% and 53%,
respectively. The comparable numbers for 2006 are 29.4% and 28.5%, respectively. Equally
interestingly, 85% and 80% of the account balance of these participants, respectively, is held
in TDFs. There seems to be a demonstrable trend towards increasing use of TDFs by younger
workers. If investment choices in retirement plans are assumed to be stable, then the
expectations of a significant increase in assets of TDFs with the passage of time seems logical.

For the year 2014, ICI reports that new fund flows into indexed funds increased by 30%
in addition to a 93% increase in 2013. A significant portion of these funds was into equity
index funds. Remarking on the growth ICI concludes, “Demand for index mutual funds
remained strong in 2014.” Index funds reflect a low cost tax efficient passive investment
philosophy. TDFs are steeped in a similar belief that discourages frequent readjustment of
asset allocation and portfolio holdings. It may not be a leap to assume that an investor
embracing index funds is likely to be attracted to TDFs and vice versa. Given the trend of
increasing assets directed towards TDFs, the design of these retirement savings vehicles becomes
important to retirement well-being of a significant proportion of the working population.

3. Investment theory and glide paths

Towers Watson (2010) reports

“… a 2009 Watson Wyatt research study found that employees categorized as ‘intermediate/
long-horizon’ investors planning to retire in 15 years or more experienced typical median
declines of 27% to 37% in the value of their plans. This magnitude of loss was attributable
to the large exposure to equity investments (51% to 95% across 2020–2055 target horizon
funds evaluated) considered appropriate for investors focused on wealth accumulation. What
is troubling, however, is that ‘short-horizon’ investors planning to retire in two to seven years
(2010–2015 target horizon funds) suffered a median loss of 21%, and as high as 36% for
some funds. These individuals have limited ability to recover from the losses incurred other
than by deferring retirement and significantly increasing their rate of retirement saving
(p. 1).”

D’Antona (2008) aptly highlights a similar plight of TDF owners retiring in the period
2008–2010. She also points to the wide range of losses suffered by such investors because
of deficient glide paths. The glide path of TDFs has a significant influence on investment
outcomes and consequently on retirement well-being. A fund’s glide path primarily refers to
the manner in which the fund asset allocation changes over time until the target date is
reached or beyond. The evidence on the efficacy of glide path design is discussed next.

Gomes, Kotlikoff, and Viceira (2008), study asset allocation in the context of a variable
labor supply. They propose that the ability of individuals who do poorly in financial markets
can make up the performance shortfall by working more. They demonstrate that the welfare
loss is negligible when using a typical life cycle mutual fund. Comparatively, there is a

240 S. Singh / Financial Services Review 25 (2016) 235–262



significant loss of welfare in investing in stable value funds. They argue in favor of higher
equity exposures during retirement as the presence of bond like cash flows in the form of
pension benefits provide better risk taking capabilities. Pfau and Kitces (2014) further
endorse the idea of upward sloping glide paths during retirement, that is, those that increase
equity exposure during retirement. Using Monte Carlo simulations, they demonstrate that
glide paths that increase equities from 30% to 60% during retirement reduce the probability
and magnitude of run outs as compared with glide paths that do the opposite that is, reduce
equity exposure from 60% to 30% during retirement. Later Kitces and Pfau (2015) combine
a rising equity allocation glide path with a dynamic asset allocation scheme. In this scheme,
equity exposures are increased when equity markets are deemed undervalued and reduced
when overvalued. Using actual return data and overlapping periods, they demonstrate that a
rising glide path is most suited in overvalued markets. Blanchett (2015) points to the fact that
the initial condition of financial markets is often ignored in many asset allocation decisions
as one is embarking upon, or is close to retirement. Utilizing a range of varying market
conditions, he concludes that upward sloping glide paths representing increasing allocations
to equities seem to perform well in relatively higher return environments and higher
allocations to lower risk assets seem to do well in low return markets.

Basu and Drew (2009) are perhaps the first to question the conventional wisdom of
reducing equity allocation in TDF portfolios as one nears the target date. They point to the
fact that individuals contribute savings to the retirement portfolio on a regular basis and the
sum total of the initial portfolio increases with time even if one assumes a rate of return of
zero on initial portfolio and contributions. Reducing equity allocation in such portfolios at a
point in time close to the withdrawal phase is detrimental to portfolio size at terminal date.
The difference between the lifecycle and contrarian strategy (one that increases equity
allocation closer to terminal date) is mostly driven by performance close to terminal years.
The authors do acknowledge the chance of ruin increase under such a strategy and that the
behavioral aspects of such a strategy cannot be ignored.

Pfau (2010) uses Monte Carlo simulations to illustrate that investors with modest risk
aversion will find traditional glide paths more suitable as compared with a fixed allocation.
He points to the failure to consider expected utility and, therefore, comparative risk aversion
in contrarian glide paths of some studies. Contrary to Basu and Drew, he shows that higher
utility is derived from traditional life cycle strategies when the investor wishes to avoid
probability of ruin.

Capitalizing further on the earlier findings, Basu, Bryne, and Drew (2011) also suggest a
dynamic asset allocation strategy predicated on portfolio value rather than the age of the
participant. The key to such a strategy would be the setting of an accumulation rate and this
accumulation rate guides the asset allocation over time. The authors utilize simulations to
demonstrate the benefits of such a strategy. Yoon (2010) suggests a risk budget based
methodology to designing glide paths. Under such a system, periodic adjustments to asset
allocation are driven by the available risk budget. The advantage of such a strategy as Yoon
points out is that the asset allocation stays within the risk tolerance at all times. The downside
of such a system is the complexity of the system and the inability to maximize returns.

Spitzer and Singh (2011) study the glide path of 22 fund families and classify glide paths
utilized by these fund families in two broad categories, “Late Descent” or “Early Descent.”

241S. Singh / Financial Services Review 25 (2016) 235–262



In late descent glide paths, the percentage of stocks stays constant at a high proportion for
the first part of the life cycle and then begins to fall rapidly over the latter part of the
investment period. In “Early Descent” the asset allocation begins with a high proportion in
equities and, the percentage of stocks falls gradually over the entire holding period. Both
strategies are shown to be inferior to a static allocation. The Investment Company Institute
(2014) uses three similar classifications to illustrate glide paths. The first, “allocates 50% of
its assets to equity as of the target date (‘RET’) and reaches its most conservative allocation
(20% equity) 15 years later.” In the second, a constant proportion is maintained for the first
20 years and then the asset allocation glides to the most conservative allocation five years
before the retirement date. In the third set of glide paths, a constant allocation is maintained
in the first 20 years and the asset allocation is gradually reduced to more conservative
through 30 years into retirement. Even with same securities in all portfolios in the preceding
strategies, the investment outcomes are likely to be different because of asset allocation and
its adjustment. The other factor affecting the investment outcome in these funds is the
investment strategy and security selection in the portfolio. The primary distinction is between
the uses of actively and passively managed funds in the lineup. Findings of the ICI study
broadly confirm results reported by similar studies related to glide path efficacy.

Arnott, Sherard, and Wu (2013) are critical of traditional glide paths. If the goal is to
maximize wealth accumulation during earning years and minimize longevity and lifestyle
risk during retirement, they argue that traditional glide paths fail on both accounts. The
reasons for the failure are suboptimal asset class exposure, inefficient risk and return balance
including lack of diversification and an assumption of constant risk premiums. Estrada
(2014) validates their findings in a global context through analysis of returns of 19 countries
covering a period of 110 years. He finds that ten alternate strategies that include five different
contrarian and five static equity allocations with varying holding periods, tend to outperform
traditional glide paths. The interesting query raised by both sets of authors is what constitutes
risk for an individual investor. By traditional measure of volatility as a proxy for risk, TDFs
tend to provide greater certainty of outcomes but as shown many times before, other
strategies tend to provide higher end of period wealth accumulation and lower run outs
(longevity risk). Idzorek, Stempien, and Voris (2013) provide a Glide Path Stability Score
(GPSS) to each fund family offering TDFs. They calculate this score by comparing the
consistency in asset allocation of different vintage TDFs from the same family. A lower
glide path stability score indicates that the glide paths of different vintages of the same family
have more variability in asset allocation for the same time remaining to retirement. They
conclude, “while glide path changes are not necessarily bad, we believe that unannounced
and unjustified changes in glide paths should be viewed with extreme scrutiny, given that
investors and sponsors select these investments based on expectations of risk (p. 81).”

Optimal glide paths during retirement have also received some attention from researchers.
Spitzer and Singh (2008) use a bootstrap simulation to study the shortfall risk of TDFs during
the retirement years. Shortfall risk is defined as the probability of running out of money
during retirement. They classify TDFs into three types of glide paths: Steep, Gentle, and
Fixed 25/75. They show that all three glide path strategies have higher shortfall risk than a
constant 50/50 allocation. They urge the designers of target-date mutual funds to “rethink
their asset allocation during retirement (p. 151).” Kalman (2011) uses a similar bootstrap

242 S. Singh / Financial Services Review 25 (2016) 235–262



methodology to confirm the findings of Spitzer and Singh. He also demonstrates that after
accounting for equity risk premium, a bond heavy portfolio in retirement reduces the
probability of run outs.

The question of whether glide paths should end at retirement or extend through retirement
remains unresolved. Clark and Hood (2009) recognize the need for maintaining a constant
real withdrawal rate during retirement. To that end, they focus on designing TDFs that are
especially suitable for the withdrawal phase of the life cycle. Such TDFs will allocate higher
and a constant proportion to riskier assets to accomplish the goal of constant real withdrawals
during retirement.

This brings up a worthwhile question: How is the choice of various glide paths supported
by investment theory? Two broad inter-related theoretical constructs are used to examine this
question: (1) Life-cycle investing incorporating human capital and, (2) Liability driven
investing. These are often referred to frequently in the context of glide paths utilized by
TDFs. A discussion of each theory in the context of glide paths is presented next. Both lines
of examination assume an underlying adherence to mean-variance optimization and portfolio
theory constructs, albeit with constraints in both cases.

3.1. Life-cycle investing incorporating human capital

The life cycle theory of consumption and portfolio choice posits that at any point in time
in a human’s life, one’s endowment consists of financial wealth and human capital. The
human capital is the present value of lifetime earnings, which is stochastic. This variability
in earnings results from controllable and uncontrollable factors. Increasing human capital
through skills and education and the ability to work the number of hours chosen along with
capacity to retire early can be treated as controllable factors. Uncontrollable factors include
being forced to work fewer than intended hours or not being able to work at all because of
unemployment, sickness or premature death. Over a lifetime, an individual converts human
capital into financial capital and it is assumed that one has exhausted all human capital at
retirement. For a detailed exposition, see Mayers (1972), Williams (1978) and Merton
(1969).

As a practical application of this theory, financial advisors usually look at a human life as
compilation of four consecutive stages: accumulation, consolidation, withdrawal, and gifting.
The first stage starts when the individual has accumulated some human capital through
training and/or education and is ready to start the conversion of this capital into financial
capital. Most analysis and research assumes an initial endowment of zero for financial capital
at the start of this stage of the life cycle. This is also perhaps the longest of the four stages
of the life cycle that starts at around 22 to 25 years of age and continues to about 10 years
before retirement. The consolidation phase usually follows the accumulation phase. In this
phase, the preparation for retirement begins. The primary activity during this stage is the
transfer of financial capital to less risky investments, since there is a diminished ability to
recover from a significant loss of portfolio value because of a shortened time horizon. After
the consolidation phase, the depletion of the accumulated financial capital starts in the
withdrawal phase. The final stage of the life cycle is the gifting phase where one plans for
the inevitable.

243S. Singh / Financial Services Review 25 (2016) 235–262



While a significant number of TDFs broadly adhere to the preceding life cycle framework,
there is a wide disparity in terms of period allocated to each stage of the cycle. While some
funds move the asset allocation to bonds quite early in the cycle, others maintain a constant
exposure to fixed income assets over extended periods. Research themed around human
capital and life cycle investing relevant to TDFs is discussed next.

Viciera (2009) describes a case where an individual has a stable job that implies steady
cash inflows. The employment prospects and size of compensation are also not highly
correlated with the performance of the stock market. These resemble cash flows from a bond
more than common stock. This implies that the investment portfolio of such an individual
should hold a high proportion of stocks. As one gets closer to retirement, the bond like
component of one’s portfolio (human capital) is depleted. This creates the need to increase
actual bonds in this integrated portfolio. Viceira also indicates that in cases where human
capital is volatile and strongly positively correlated to the stock market, equity exposure
should be limited. Finally, since expected returns change over long holding periods, occa-
sional adjustments to glide paths may be useful. He concludes that target date investing is
consistent with the human capital approach.

For long holding periods Viciera (2009) finds that having a TDF in a pension plan as
compared to a money market fund alone results in higher utility for the participant. Another
suggestion for better TDF design implies a significant exposure to real assets in the asset
allocation. Funds with higher allocation to TIPS, therefore, are superior in design in Viciera’s
view.

With respect to accumulation years, Shiller (2005) makes some interesting observations.
The optimal asset allocation between human and equity capital is dependent on the correlation
between the returns of these two assets. He notes that these correlation estimates vary over a
broad range. He also points to recommended equity exposures as a proportion of portfolio value
in the literature, of as low as 20% to as high as 300% for young workers depending on the
assumptions made regarding the relationship of human capital and equity returns.

A comprehensive examination of the influence of labor income on the portfolio decision
is provided by Cocco et al. (2005). This provides a good basis for a discussion on the optimal
design of glide paths as related to TDFs. In the presence of labor income that has low
correlation to equity returns they find support for decreasing proportion of equity investments
as one ages. An interesting find was that the most significant welfare loss of about 2% per
year occurs when one ignores labor income in the portfolio decision. The loss is less
significant when one ignores labor income risk only. The most significant loss in welfare
arises from “disastrous labor income realization,” meaning unemployment or disability
unsupported by safety nets. In light of the findings, glide paths that have declining equity
exposures over time, that incorporate nontraded labor income in the asset allocation choices
can result in higher welfare outcomes.

Bodie and Treussard (2007) provide additional insight into the role of riskiness of human
capital in determining an optimal glide path. They classify certain individuals as “natural
TDF holders.” These are individuals with risky human capital, that is, their human capital
betas are high. Such individuals will experience substantial welfare gains if offered a
relatively safe TDF. To reflect the conversion of an individual’s portfolio from human capital

244 S. Singh / Financial Services Review 25 (2016) 235–262



to financial capital and the translation of human capital from “stock like” to “bond like,” the
glide path should be “humped”; not the traditional linear reduction with age.

Boscalajon (2011) provides a discussion of “critical wealth,” a point where financial and
human capital is equal. He argues that it is from this point on that a systematic transfer to less
risky financial assets should start taking place. Utilizing Monte Carlo simulations and utility
functions that have a coefficient of risk aversions ranging from 1 to 10, Pfau (2011) finds
support for glide paths that adhere to traditional strategy of high initial equity allocation and
declining risky assets over time as utility maximizing.

As mentioned earlier, the value of human capital is the present value of expected future
earnings. Using Bureau of Economic Analysis (BEA) data and yields on investment grade
corporate bonds in various industries, Blanchett and Straehl (2015) attempt to quantify
human capital in 12 industries. The authors treat the estimate of the return and risk of human
capital across industries as a separate asset class. These then constitute inputs in a portfolio
optimization scheme along with other traditional asset class returns. As might be expected,
they find that “the optimal equity allocation decreases with age, riskier employment, and
riskier homeownership, whereas it increases with guaranteed pension income (p.1).”

Another and more recent stream of thought affecting glide path design is liability driven
investing (LDI). It is derived from defined benefit pension plan design. Under such a strategy,
the determination of retirement liabilities is first made and then a glide path is designed to
optimally meet them. The body of knowledge as related to glide path design motivated by
LDI is discussed next.

3.2. Liability driven investing

LDI derives its inspiration from the management of assets to meet future liabilities, akin
to the management of defined benefit plan assets. Examples of prominent practitioners of
LDI are defined-benefit plan sponsors and insurance companies. Traditional portfolio theory
assumes that the optimal portfolio is independent of investor’s risk preferences Markowitz
(1952), Sharpe (1964). Risk aversion is incorporated through the proportion of risk free
holdings of the investor. LDI on the other hand, focuses on designing portfolios to meet
future liabilities. It is an individual exercise for retirement portfolio design and management
as each individual liability set is different. A defined-contribution plan can be managed
similarly if one considers retirement spending needs as a liability. Under LDI, for TDFs the
goal in building a retirement portfolio, is not a maximization of portfolio expected returns
given a level of risk, but rather to ensure that the portfolio has sufficient assets to support
retirement withdrawals and that it does not run out of money during the lifetime of the
investor. Minimizing longevity risk is the priority under such a strategy.

The goal is to attain a set of real cash flows that will last with certainty throughout
retirement. Thus, inflation protection during retirement years is an embedded objective of all
glide paths in a LDI based framework. In the design of glide paths, attaining this goal often
takes precedence over portfolio optimization.

According to Meder (2012), “the first order asset allocation decision is no longer focused
on the split between equities and core fixed income but focused on deriving the split between
a return-seeking asset (RSA) and a liability hedging assets (LHA) component. The RSA

245S. Singh / Financial Services Review 25 (2016) 235–262



component seeks to generate returns in excess of the expected liability return (growth in the
present value of the liability attributable to the passage of time), similar to the discount rate
on the liability (p. 117).” While the preceding comment is in the context of a defined benefit
pension, it provides a unique manner of looking at glide paths of TDFs. It provides a dual
goal in the accumulation phase of the investing horizon; first, to hedge the anticipated
liability and the second, to provide growth to the portfolio.

Idzorek (2008) provides a detailed discussion on designing TDFs that incorporate liability
relative portfolio optimization. If an individual’s retirement expenses can be thought of as a
set of real cash flow liabilities, then a retiree’s portfolio during the saving years is optimized
subject to this liability constraint. This liability constraint is implemented through a short
position in a portfolio of TIPS. Idzorek provides a comparative visual for such optimization
vis-a-vis an asset only optimization. The significant difference is the heavy relative over-
weighting in real assets like TIPS and commodities in such portfolios that are subject to the
liability constraint of a short position in TIPS.

Another advantage of liability relative investing is the determination of the fund status.
This is analogous to an underfunded or over funded status of a defined benefit plan. If a
retiree finds early on in their life of the underfunded status, and they are responsible about
it, then there may be an opportunity to alter their savings-consumption mix accordingly.

There seems to be sufficient support for designing glide paths that embrace meeting
retirement liabilities as opposed to optimizing asset only portfolios as the primary goal. It
may be possible to jointly accomplish both asset only optimization and LDI, but in the
presence of constraints, for example long only portfolios, the task becomes quite challenging.
Additionally, optimization processes are extremely sensitive to the value of inputs, for
example, a small change in volatility causes a significant shift in the optimal portfolio. In
light of the preceding, a LDI only based approach, where portfolio optimization is a
secondary constraint to designing glide paths might be worth pursuing.

Whitten and Thuerbach (2015) provide an illustration of using LDI driven glide paths. The
goal is to provide a level of real income during retirement. Using a glide path that has a
higher proportion of traditional inflation hedge assets (TIPS � Commodities � Real estate)
and lower proportion of traditional equities over time, through Monte Carlo simulations, they
demonstrate that such glide paths provide narrower distributions and lower Value at Risk
(VAR) across all retirement horizons. It is important to point out that the analysis is to
retirement and does not extend through retirement.

Anecdotally, it seems that LDI is garnering increasing interest of practitioners. In addition
to PIMCO, Dimensional Funds Advisors (2016), a mutual fund company, recently intro-
duced TDFs of various vintages premised on LDI. A review of the promotional literature and
prospectus of these funds provide a good example of application of the ideas discussed
previously in this section. For example, the 2005, 2010, and 2015 vintages at the end of May
31, 2016 held about two-thirds of their assets in TIPS. This is a good illustration of
implementation of the recommendations of Whitten and Thuerbach (2015). A summary of
the research on the human capital aspects and liability driven investing as pertaining to TDFs
appears in Table 2.

For comparative evaluation and justifying changes to existing TDF design, it becomes
important to measure outcomes from TDF investing. TDFs have now been in existence for

246 S. Singh / Financial Services Review 25 (2016) 235–262



T
ab

le
2

T
ar

ge
t-

da
te

fu
nd

re
se

ar
ch

su
m

m
ar

y

A
ut

ho
r(

s)
Y

ea
r

T
op

ic
fo

cu
s

Pu
bl

ic
at

io
n

K
ey

fin
di

ng
/c

on
tr

ib
ut

io
n

A
gn

ew
et

al
.

20
03

A
na

ly
si

s
of

40
1(

K
)

po
rt

fo
lio

s
A

m
er

ic
an

E
co

no
m

ic
R

ev
ie

w
Pa

rt
ic

ip
an

t
as

se
t

al
lo

ca
tio

n
ne

gl
ec

t
an

d
ra

re
re

ba
la

nc
in

g.
T

an
g

et
al

.
20

10
Jo

ur
na

l
of

Pu
bl

ic
E

co
n.

O
pt

im
al

po
rt

fo
lio

of
fe

ri
ng

s
un

do
ne

by
po

or
pa

rt
ic

ip
an

t
ch

oi
ce

.
L

an
ds

be
rg

20
07

T
D

F
m

ar
ke

t
an

d
re

gu
la

to
ry

en
vi

ro
nm

en
t

J
of

D
ef

er
re

d
C

om
p.

E
nr

ol
m

en
t

pr
oc

ed
ur

es
an

d
re

g.
re

lie
f

pr
om

ot
e

as
se

t
gr

ow
th

.
L

an
ds

be
rg

20
14

J
of

P.
Pl

an
.

&
C

om
pl

ia
nc

e
Fi

du
ci

ar
y

re
sp

on
si

bi
lit

y
re

la
te

d
to

T
D

Fs
fo

r
pl

an
sp

on
so

rs
.

R
ic

ha
rd

s
et

al
.

20
14

T
IA

A
-C

R
E

F
D

ia
lo

gu
e

T
D

Fs
ga

rn
er

in
g

m
os

t
re

ce
nt

re
tir

em
en

t
co

nt
ri

bu
tio

ns
.

B
al

du
zz

i
et

al
.

20
15

N
B

E
R

W
or

ki
ng

Pa
pe

r
Si

gn
ifi

ca
nt

ri
sk

ta
ki

ng
di

ve
rs

ity
in

lo
w

er
m

ar
ke

t
sh

ar
e

fu
nd

s.
B

as
u

&
D

re
w

20
09

G
lid

e
pa

th
de

si
gn

J
of

Po
rt

fo
lio

M
an

ag
em

en
t

Pe
rf

or
m

an
ce

in
la

te
ye

ar
s

do
m

in
at

es
re

tir
em

en
t

ac
cu

m
ul

at
io

n.
C

la
rk

&
H

oo
d

20
09

Jo
ur

na
l

of
In

ve
st

in
g

E
qu

iti
es

to
in

cr
ea

se
in

re
tir

em
en

t
fo

r
co

ns
ta

nt
re

al
re

tu
rn

s.
Pf

au
20

10
Fi

na
nc

ia
l

Se
rv

ic
es

R
ev

ie
w

L
if

e
cy

cl
e

fu
nd

s
pr

ov
id

e
hi

gh
er

ut
ili

ty
if

ri
sk

av
er

si
on

is
m

od
es

t.
Pf

au
20

11
A

pp
lie

d
E

co
no

m
ic

L
et

te
rs

D
et

er
m

in
e

gl
id

e
pa

th
s

ba
se

d
on

co
ef

fic
ie

nt
of

ri
sk

av
er

si
on

.
Y

oo
n

20
10

J
of

A
ss

et
M

an
ag

em
en

t
R

is
k

bu
dg

et
ba

se
d

gl
id

e
pa

th
s

in
st

ea
d

of
de

cl
in

in
g

ri
sk

to
le

ra
nc

e.
B

as
u

et
al

.
20

11
J

of
Po

rt
fo

lio
M

an
ag

em
en

t
D

yn
am

ic
al

lo
ca

tio
n

gu
id

ed
by

ac
cu

m
ul

at
io

n
ra

te
in

st
ea

d
of

tim
e.

A
rn

ot
t

et
al

.
20

13
Jo

ur
na

l
of

R
et

ir
em

en
t

T
ra

di
tio

na
l

gl
id

e
pa

th
s

fa
il

to
lim

it
lif

es
ty

le
an

d
lo

ng
ev

ity
ri

sk
.

Id
zo

re
k

et
al

.
20

13
Jo

ur
na

l
of

In
ve

st
in

g
G

lid
ep

at
h

st
ab

ili
ty

sc
or

e
(G

SS
)

to
ex

am
in

e
gl

id
e

pa
th

dr
if

t.
Pf

au
&

K
itc

es
20

14
J

of
Fi

na
nc

ia
l

Pl
an

ni
ng

R
ec

om
m

en
d

in
cr

ea
si

ng
eq

ui
tie

s
du

ri
ng

re
tir

em
en

t
ye

ar
s.

B
la

nc
he

tt
20

15
J

of
Fi

na
nc

ia
l

Pl
an

ni
ng

Si
gn

ifi
ca

nc
e

of
in

iti
al

m
ar

ke
t

co
nd

iti
on

s
at

re
tir

em
en

t.
K

itc
es

&
Pf

au
20

15
J

of
Fi

na
nc

ia
l

Pl
an

ni
ng

C
on

di
tio

na
l

gl
id

e
pa

th
s

ba
se

d
on

eq
ui

ty
m

ar
ke

t
va

lu
at

io
ns

.

247S. Singh / Financial Services Review 25 (2016) 235–262



T
ab

le
2

(C
on

tin
ue

d)

A
ut

ho
r(

s)
Y

ea
r

T
op

ic
fo

cu
s

Pu
bl

ic
at

io
n

K
ey

fin
di

ng
/c

on
tr

ib
ut

io
n

N
ag

en
ga

st
,

B
uc

ci
,

&
C

oa
ke

r
20

06
T

D
F

pe
rf

or
m

an
ce

ev
al

ua
tio

n
Po

pp
in

g
th

e
H

oo
d

T
D

Fs
la

ck
im

ag
in

at
io

n
an

d
ca

nn
ot

ou
tp

er
fo

rm
m

ar
ke

t.
Su

rz
&

Is
ra

ls
en

20
07

J
of

Pe
rf

.
M

ea
su

re
m

en
t

T
D

Fs
fa

il
to

ou
tp

er
fo

rm
Pu

re
T

ar
ge

t-
da

te
In

de
xe

s.
Sp

itz
er

&
Si

ng
h

20
08

Fi
na

nc
ia

l
Se

rv
ic

es
R

ev
ie

w
T

ra
di

tio
na

l
gl

id
e

pa
th

s
la

g
co

ns
ta

nt
m

ix
po

rt
fo

lio
.

B
la

nc
he

tt
et

al
.

20
10

J
of

Pe
ns

io
n

B
en

efi
ts

In
co

rp
or

at
in

g
ri

sk
lo

w
er

s
di

sp
er

si
on

an
d

re
tir

em
en

t
fa

ilu
re

.
D

ol
vi

n
et

al
.

20
10

J
of

Fi
na

nc
ia

l
Pl

an
ni

ng
G

lid
e

pa
th

s
ou

tc
om

es
ar

e
eq

ui
va

le
nt

to
st

at
ic

as
se

t
al

lo
ca

tio
ns

.
K

al
m

an
20

11
Fi

na
nc

ia
l

Se
rv

ic
es

R
ev

ie
w

B
on

d
he

av
y

ri
sk

pr
em

iu
m

ad
ju

st
ed

po
rt

fo
lio

s
re

du
ce

ru
n

ou
ts

.
L

ip
to

n
&

K
is

h
20

11
Fi

na
nc

ia
l

Se
rv

ic
es

R
ev

ie
w

L
if

e
cy

cl
e

fu
nd

s
ad

d
lit

tle
va

lu
e

on
ri

sk
-

ad
ju

st
ed

ba
si

s.
B

ar
e

&
G

re
ve

s
20

13
V

ie
w

po
in

t
R

us
se

ll
R

es
.

Fo
cu

s
on

re
tu

rn
so

ur
ce

s
an

d
de

-e
m

ph
as

iz
e

re
la

tiv
e

pe
rf

or
m

an
ce

.
C

as
si

dy
et

al
.

20
14

J
of

Po
rt

fo
lio

M
an

ag
em

en
t

W
id

el
y

di
sc

lo
se

d
be

nc
hm

ar
k

en
ab

le
s

in
st

itu
tio

na
l

be
st

pr
ac

tic
es

.
C

ha
ng

et
al

.
20

14
Fi

na
nc

ia
l

Se
rv

ic
es

R
ev

ie
w

T
D

Fs
pr

ov
id

e
be

tte
r

w
el

fa
re

w
he

n
co

m
pa

re
d

to
lif

e
cy

cl
e

fu
nd

s.
E

st
ra

da
20

14
J

of
Po

rt
fo

lio
M

an
ag

em
en

t
C

on
fir

m
s

A
rn

ot
t

et
al

.,
U

S
fin

di
ng

s
fo

r
19

ot
he

r
co

un
tr

ie
s.

T
an

g
&

L
in

20
15

J
of

A
ss

et
M

an
ag

em
en

t
In

ap
pr

op
ri

at
e

be
nc

hm
ar

k
ch

oi
ce

re
su

lts
in

w
el

fa
re

lo
ss

of
67

%
.

W
ill

ia
m

s
19

78
In

co
rp

or
at

in
g

hu
m

an
ca

pi
ta

l
th

eo
ry

in
T

D
Fs

Jo
ur

na
l

of
B

us
in

es
s

In
iti

al
pa

pe
r

on
in

co
rp

or
at

in
g

hu
m

an
ca

pi
ta

l
in

po
rt

fo
lio

ch
oi

ce
.

C
oc

co
et

al
.

20
05

R
ev

.
of

Fi
na

nc
ia

l
St

ud
ie

s
Ig

no
ri

ng
hu

m
an

ca
pi

ta
l

re
su

lts
in

a
2%

w
el

fa
re

lo
ss

pe
r

ye
ar

.
Sc

hi
lle

r
20

05
T

he
E

co
no

m
is

t’
s

V
oi

ce
H

um
an

an
d

fin
an

ci
al

ca
pi

ta
l

re
tu

rn
s

co
rr

el
at

io
n

ef
fe

ct
.

B
od

ie
et

al
.

20
07

Fi
na

nc
ia

l
A

na
ly

st
s

Jo
ur

na
l

R
ec

om
m

en
d

hu
m

pe
d

gl
id

e
pa

th
s

to
in

co
rp

or
at

e
hu

m
an

ca
pi

ta
l.

G
om

es
et

al
.

20
08

A
m

er
ic

an
E

co
n.

R
ev

ie
w

A
ss

et
al

lo
ca

tio
n

in
th

e
pr

es
en

ce
of

va
ri

ab
le

la
bo

r
su

pp
ly

.

248 S. Singh / Financial Services Review 25 (2016) 235–262



T
ab

le
2

(C
on

tin
ue

d)

A
ut

ho
r(

s)
Y

ea
r

T
op

ic
fo

cu
s

Pu
bl

ic
at

io
n

K
ey

fin
di

ng
/c

on
tr

ib
ut

io
n

V
ic

ei
ra

20
09

L
if

e
C

yc
le

Fu
nd

s
(C

ha
pt

er
)

H
um

an
ca

pi
ta

l
is

bo
nd

lik
e

so
m

or
e

bo
nd

s
cl

os
er

to
re

tir
em

en
t.

B
os

ca
ljo

n
20

11
Fi

na
nc

ia
l

Se
rv

ic
es

R
ev

ie
w

In
tr

od
uc

es
th

e
co

nc
ep

t
of

cr
iti

ca
l

w
ea

lth
th

at
af

fe
ct

s
al

lo
ca

tio
n.

B
la

nc
he

tt
et

al
.

20
15

Fi
na

nc
ia

l
A

na
ly

st
s

Jo
ur

na
l

Q
ua

nt
if

y
hu

m
an

ca
pi

ta
l

re
tu

rn
an

d
ri

sk
in

12
in

du
st

ri
es

.
Id

zo
re

k
20

08
Im

m
un

iz
in

g
re

tir
em

en
t

lia
bi

lit
ie

s
U

np
ub

lis
he

d
m

an
us

cr
ip

t
L

ia
bi

lit
y

co
ns

tr
ai

nt
m

od
el

ed
as

a
sh

or
t

po
si

tio
n

in
T

IP
s.

M
ed

er
20

12
Jo

ur
na

l
of

In
ve

st
in

g
Po

rt
fo

lio
sp

lit
am

on
g

lia
bi

lit
y

he
dg

in
g

an
d

re
tu

rn
se

ek
in

g
as

se
ts

.
W

hi
tte

n
et

al
.

20
15

PI
M

C
O

So
lu

tio
ns

H
ig

he
r

ra
tio

of
in

fla
tio

n
he

dg
e

as
se

ts
re

su
lt

in
lo

w
er

V
A

R
s.

249S. Singh / Financial Services Review 25 (2016) 235–262



sufficient time to provide return data for such a preliminary assessment. A discussion on
benchmarking and evaluation of TDF performance is presented next.

4. Benchmarking and performance evaluation

In this section, a discussion on two aspects of TDF performance is presented. In the first,
a framework for benchmarking and performance attribution of TDFs is discussed. In the
second, a summary of accumulated empirical evidence on the performance of TDFs is
provided. Before the discussion on performance evaluation and attribution of TDFs, an
important fact regarding TDF strategies need to be pointed out. Most TDF strategies assume
a long-term investment horizon. Many TDFs, for example, those of 2050 or 2055 vintage,
assume a holding period of 40 years or more, and this is for just “to” retirement TDFs not
“through” retirement. The long horizon becomes even more significant when one takes into
account the finding that returns in the last few years of the investment horizon are the most
significant, as pointed out by Basu et al., Therefore, any performance evaluation methodol-
ogy should take into account the assumed long-term investment horizon of the investor. It
seems that the traditional one, three, or five-year returns might not be adequate to evaluate
TDF performance.2

A significant portion of DC plan assets is TDFs, and as mentioned in the earlier part of
the paper, assets invested in TDFs are expected to grow. Monitoring duties of the plan
sponsor usually involve performance evaluation of plan offerings. Cassidy et al. (2014)
emphasize the need for a benchmark based performance evaluation in defined contribution
plans. The benchmark enables better communication between plan sponsors and investment
managers. It affords the plan participants benefits of institutional best practices. A benchmark
also lets the plan sponsor clearly define risk to the investment managers. Cassidy et al.
recommend separate benchmarks for accumulation and decumulation stages.

The subject of benchmarking and attribution of TDF performance is best understood in the
attribution framework borrowed from Bailey, Richards, and Tierney (2007). Let M be the
return that represents a neutral asset allocation and passive return for underlying asset classes
of a TDF. As applicable to TDFs the benchmark, M captures the utility function of the
investor including an appropriate degree of risk aversion. Neutral asset allocation in this
context would result in minimum acceptable welfare derived from the TDF investment
strategy.

For an actively managed TDF, a portfolio manager can add value and improve welfare,
through three broad sets of decisions. First, the manager can deviate from neutral asset class
weights and contribute value to the portfolio. For example, overweighting fixed income
securities when relative returns from this asset class are forecasted to be favorable in the view
of the manager. Let this be denoted by W. Next, they can alter allocation to sectors within
an asset class. For example, over weighting mortgage-backed securities relative to Treasur-
ies. Let this be denoted by S. Finally, security selection can be a source of value. Let this be
denoted by A. A represents over and underweighting of securities relative to the benchmark.
Assuming interaction affect between S and A to be minimal, the portfolio return P can then
be represented as,

250 S. Singh / Financial Services Review 25 (2016) 235–262



P � M � W � S � A (1)

For a TDF with underlying index funds as investment options, both S and A are required to
be 0, so the portfolio return becomes,

P � M � W (2)

In cases, where no deviation from a neutral allocation is permitted, but underlying funds are
actively managed,

P � M � S � A (3)

In cases, where no deviation from a neutral allocation is permitted, and underlying funds are
indexed,

P � M (4)

In the case of Eq. (4) the role of the manager is largely administrative as they can exercise
no discretion in asset allocation or security selection decision. A couple of additional
interesting points about the above frame work. In most cases, W is restricted to a range of
real life funds. Second, being that a TDF is essentially a fund of funds, S and A are
often beyond the control of the TDF manager and a domain of the manager of the underlying
fund.

Assuming a similar setup, Bare and Greves (2013) recommend that investment perfor-
mance and participant success be segregated in the performance evaluation of TDFs. As
participant success in accomplishing investment goals is individual and varies across
participants, the more objective endeavor would be to study investment performance. The
other recommendation made by the authors is to deemphasize peer relative performance
evaluation.

The long-term asset allocation of the fund as reflected by the fund’s glide path represents
the TDF’s strategic allocation mix. If the fund relies on active management, then there is an
additional component of security selection and tactical return if the fund mandate permits
alteration of asset allocation, perhaps within a range. The focus should be upon identifying
the sources of risk and return of the TDF portfolio, according to the authors.

Bare and Greves (2013) suggest a three-pronged approach to performance attribution.
First, developing a simple benchmark representing the growth and capital preservation
components of the fund. Second, determining a composite benchmark representing the
returns of the specific target components of the fund. For example, performance of real assets
in a neutral allocation. Together, this is analogous to M in Eq. (1). Finally, a return
comparison between composite and simple benchmark, and then between the fund return and
the simple benchmark to isolate the value added by the manager through security selection
and tactical choices. This is equivalent to W�S�A in Eq. (4). With this framework in place,
it is instructive to examine how the benchmarking tools available stand up to fundamental
principles of a good benchmark. A number of index providers have created target date
indexes for benchmarking and performance evaluation. According to Bailey, Richards, and
Tierney (2007), a valid benchmark should possess the following properties: it is (1) unam-

251S. Singh / Financial Services Review 25 (2016) 235–262



biguous, (2) measurable, (3) appropriate, (4) reflective of current investment opinions, (5)
prescribed in advance, (6) investable, and (7) accountable.

At present, there are multiple target-date indexes available for benchmarking TDF per-
formance. Descriptive literature provided by each index provider highlights the superiority
of each index design. A summary of these indexes appears in Table 3. Please note that
despite reasonable efforts, the author of this paper is unable to find up to date information on
some of these indexes. In the table, the prominent feature of each benchmark is highlighted.
To provide the reader with an academic assessment of index qualities rather than compar-
ative positives and negatives, an aggregate analysis based on Bailey, Richards, and Tierney’s
seven properties of a valid benchmark are presented.

As can be observed from Table 3, a wide array of investment philosophies is found in the
design of target date indexes. There are many different glide paths used by TDFs. Similarly,
there are wide and varying approaches to the underlying glide paths tracked by the indexes.
Note that the tracking error of any TDF will likely be different by benchmark, given the wide
divergence in the underlying glide path of each index. From the perspective of plan sponsors,
there seem to be sufficient historical returns to allow back testing and to determine which
index best fits their preferred glide path.

All indexes shown in Table 3 take into account the need for asset protection in their design
and some aim to benchmark real returns close to and during retirement. This recognizes the
need to incorporate investor risk tolerance and volatility of human capital related cash flows.
A couple of index sponsors have started offering more than one index series for the same
target date. The choice of a particular series incorporates investor risk preferences to some
extent in the glide path decision.

A word of caution regarding the calculation of returns, index rebalancing and reconsti-
tution might be appropriate. Some indexes report gross returns, others report returns net of
fees, while others report both. The influence on tracking error from this difference in
reporting returns should be recognized. All index returns are calculated from underlying
indices that have their own rebalancing and reconstitution rules. The impact on returns of this
fact is not widely understood or has been investigated by researchers. A wide variation in the
frequency of reconstitution and rebalancing of the indexes is revealed in Table 3. Time
between reconstitution and rebalancing can range from one month to one year. Too frequent
rebalancing can be expensive and may not provide sufficient time to benefit from trending
markets, while a period too long between rebalancing may lead to significant deviations from
target asset allocations for the portfolio. Again, this has not been analyzed nor there is a
consensus among researchers on an optimal time between rebalancing and reconstitution.

With respect to the seven properties of a valid benchmark, none of the indexes meets all
seven. The efficacy of target date indexes is undermined by the fact that they are indexes
derived from other indexes. All six indexes are measurable and given sufficient information,
the manager whose portfolio is benchmarked can agree to be accountable to any of the six
benchmarks. Indexes based on consensus cannot be specified in advance and, therefore, are
not investable. Similarly, equal weighted indexes do not represent the underlying market and
are technically not investable. Benchmarks based on proprietary algorithms and rebalancing
methodologies that are not clearly disclosed, or understood, do not meet the unambiguous
criteria. Arguably, a benchmark that attempts to incorporate investor risk preferences is more

252 S. Singh / Financial Services Review 25 (2016) 235–262



T
ab

le
3

T
ar

ge
t-

da
te

fu
nd

in
de

xe
s

fe
at

ur
es

In
de

xe
s

C
ha

ra
ct

er
is

tic
sa

C
al

la
n

D
ow

Jo
ne

s
R

ea
l

R
et

ur
n

D
ow

Jo
ne

s
G

lo
ba

l
D

ow
Jo

ne
s

U
.S

.
M

or
ni

ng
st

ar
L

if
et

im
e

S&
P

T
ar

ge
t

D
at

e
In

de
x

T
D

A
O

n
T

ar
ge

t
In

de
xe

s

R
is

k
pr

ofi
le

va
ri

et
y

1
1

1
1

3
1

4
G

lid
e

pa
th

s
C

on
se

ns
us

dr
iv

en
C

os
in

e
cu

rv
e

ba
se

d
C

os
in

e
cu

rv
e

ba
se

d
C

os
in

e
cu

rv
e

ba
se

d
H

um
an

ca
pi

ta
l

an
d

lia
bi

lit
y

dr
iv

en

C
on

se
ns

us
dr

iv
en

L
os

s
pr

ob
ab

ili
ty

dr
iv

en
m

od
el

A
ss

et
cl

as
se

sb
4

4
3

3
4

4
4

In
fla

tio
n

pr
ot

ec
tio

n
R

E
IT

S,
C

om
m

od
iti

es
,

an
d

T
IP

S
R

E
IT

S,
C

om
m

od
iti

es
,

an
d

T
IP

S
N

o
sp

ec
ifi

c
al

lo
ca

tio
n.

T
hr

ou
gh

un
de

rl
yi

ng
in

de
xe

s.

N
o

sp
ec

ifi
c

al
lo

ca
tio

n.
T

hr
ou

gh
un

de
rl

yi
ng

in
de

xe
s.

T
IP

S
an

d
C

om
m

od
iti

es
R

E
IT

S,
C

om
m

od
iti

es
,

an
d

T
IP

S
R

E
IT

S,
C

om
m

od
iti

es
,

an
d

T
IP

S

R
eb

al
an

ci
ng

A
nn

ua
l

Se
m

i-
an

nu
al

M
on

th
ly

M
on

th
ly

Q
ua

rt
er

ly
M

on
th

ly
M

on
th

ly
In

de
x

re
co

ns
tit

ut
io

n
A

nn
ua

l
Se

m
i-

an
nu

al
M

on
th

ly
M

on
th

ly
A

nn
ua

l
A

nn
ua

l
M

on
th

ly
U

nd
er

ly
in

g
in

de
xe

s
m

et
ho

do
lo

gy
V

ar
io

us
E

qu
al

w
ei

gh
te

d
E

qu
al

w
ei

gh
te

d
E

qu
al

w
ei

gh
te

d
C

ap
w

ei
gh

te
dc

C
ap

w
ei

gh
te

d
V

ar
io

us

U
nd

er
ly

in
g

in
de

xe
s

re
co

ns
tit

ut
io

n
m

et
ho

d

V
ar

io
us

Pr
op

ri
et

ar
y

Pr
op

ri
et

ar
y

Pr
op

ri
et

ar
y

Pr
op

ri
et

ar
y

C
om

m
itt

ee
V

ar
io

us

U
nd

er
ly

in
g

co
m

po
ne

nt
s

V
ar

io
us

D
ow

Jo
ne

s
In

de
xe

s
D

ow
Jo

ne
s

In
de

xe
s

D
ow

Jo
ne

s
In

de
xe

s
M

or
ni

ng
st

ar
In

de
xe

s
an

d
Fu

tu
re

s

E
T

Fs
an

d
C

om
m

od
ity

T
ru

st
M

ut
ua

l
Fu

nd
s

an
d

E
T

Fs

R
et

ur
ns

ca
lc

.
st

ar
t

da
te

Ja
nu

ar
y

20
06

D
ec

em
be

r
19

99
Ja

nu
ar

y
19

83
Ja

nu
ar

y
19

83
D

ec
em

be
r

20
01

M
ay

20
07

Ja
nu

ar
y

19
98

N
ot

e:
a
T

ab
le

cr
ea

te
d

fr
om

in
fo

rm
at

io
n

pr
ov

id
ed

in
M

or
ni

ng
st

ar
,

S&
P,

an
d

C
al

la
n

A
ss

oc
ia

te
s

do
cu

m
en

ts
.

b
A

ss
et

cl
as

se
s

ar
e

cl
as

si
fie

d
as

E
qu

ity
,

Fi
xe

d
In

co
m

e,
In

fla
tio

n
H

ed
ge

A
ss

et
s,

an
d

C
as

h.
In

de
xe

s
w

ith
th

re
e

as
se

t
cl

as
se

s
do

no
t

in
cl

ud
e

in
fla

tio
n

he
dg

e
as

se
ts

se
pa

ra
te

ly
.

c
C

ap
w

ei
gh

te
d

in
de

xe
s

ar
e

flo
at

ad
ju

st
ed

.
C

om
m

od
ity

ex
po

su
re

is
pr

od
uc

tio
n

w
ei

gh
te

d.

253S. Singh / Financial Services Review 25 (2016) 235–262



appropriate than the one that relies upon the investor holding suitable levels of cash to
incorporate risk tolerance. Two of the six indexes attempt this, and, therefore, seem better
designed on the appropriateness factor. Blanchett and Kasten (2010) provide initial support
for incorporating risk aversion in index design. They demonstrate that using three broad
categories of risk and dynamically adjusting allocation provide lower dispersion of returns
and more certain outcomes. Therefore, a risk appropriate benchmark may capture a neutral
welfare maximizing allocation more appropriately.

A proportion of TDFs are supported by underlying indexed funds and ETFs that are
actively managed. Sometimes fund managers have latitude regarding asset allocation in
many TDFs. The property of the benchmark being reflective of current investment opinion
becomes significant. A manager who has discretion in terms of asset allocation and security
selection should understand factors that affect the benchmark and have an opinion with
regard to the factors. A majority of TDF strategies is executed through indexed portfolios and
predetermined glide paths. Under such circumstances, the property of the benchmark being
reflective of current investment opinion is not so relevant, as manager discretion is limited.
Based on the evaluation of the seven benchmarks, once the underlying indexes and their
proportions are disclosed, all of the indexes meet the investment opinion criterion.

At best, TDF benchmark design is in its infancy and plan sponsors can expect more
refinements in time to come. Once an appropriate benchmark is selected, it becomes relevant
to measure the performance in context of the benchmark. To this point, most performance
evaluation has relied on benchmark independent performance evaluation and mostly on
simulations.

The empirical evidence on TDF performance is presented next. Nagengast, Bucci, and
Coaker (2006) study performance of TDFs from six major fund families. They utilize a
weighted score of six metrics: structure/strategy, expenses, allocation, performance, and two
measures of risk. The authors find the returns from these funds to be in line with the
performance of the market. They observe, “the asset allocations of most of the fund families
lack imagination.”

Surz and Israelsen (2007) determine that TDFs “failed to measure up to the risk-adjusted
performance standards established by the Pure Target-date Indexes.” To arrive at this
conclusion, the authors first prescribe the characteristics possessed by an ideal TDF. They
also build four benchmarks ranging from “Defensive” to “Aggressive” depending upon the
glide path applicable to each category of TDF. They arrive at the preceding conclusion based
on the risk classification of a TDF and performance relative to the benchmark.

Borrowing performance metrics from the annuity industry, Lewis (2008) uses actual
annuity quotes to calculate the probability of attaining an income replacement ratio. This
enables one to evaluate the efficacy of a particular glide path and the shortfall risk associated
with each. In the opinion of the author, this makes comparative risk assessment of glide paths
easier and, perhaps less dependent on simulated data as in other studies.

Using bootstrap simulations, Dolvin et al., (2010) find that certain dynamic strategies that
reduce equity exposure over time are equivalent to some static strategies. They also conclude
that most glide paths follow a 120-age allocation scheme, similar to a 100-age strategy.

Lipton and Kish (2011), highlight the opacity in disclosure and the wide variability in
glide paths, management fees and outcomes in life cycle investing. They construct simple

254 S. Singh / Financial Services Review 25 (2016) 235–262



Rolling Indexes based on the widely used asset allocation rule of 100-age in equities. An
evaluation of a wide range of target dates and TDFs leads them to conclude “Life Cycle
funds add little value on a risk-adjusted basis” and “that individuals with a long time horizon
may want to consider whether the apparent convenience of Life Cycle funds outweighs the
difficulties in measuring and attaining return (p. 92).”

In a study of 36 fund families, Tang and Lin (2015) find significant welfare loss from
investing arbitrarily in a TDF. They measure loss from two sources: portfolio selection and
glide path. The authors conclude that an inappropriate TDF, one whose glide path does not
reflect the risk preference of the investor can cost the investor a loss of as high as 17% in
welfare. Of this loss in welfare, the most significant contributor to welfare loss of about 67%
is attributed to inappropriate glide path choice and the remainder could be attributed to
portfolio selection. The authors advocate a risk-based TDF selection strategy. A summary of
the accumulated research on performance evaluation of TDFs appears in Table 2.

Some initial evidence on the actual performance of TDFs is now available. In a 2016
Morningstar study, Holt et al. (2016), calculate the historical 10-year annual average total
return on TDFs and compare them to various classes of mutual funds. They find that on a
total return basis the TDFs lag funds that include only equities, but do better than categories
of bond funds. There has been a tremendous flow of funds into TDFs over the past 10 years,
a point made on multiple occasions in this study. Returns calculated after taking cash flows
in and out of the portfolio might be a better gauge of the returns actually experienced by the
investor. When comparing dollar-weighted return, Holt et al., find that with the exception of
sector and diversified funds, TDFs exceeded the performance of all other categories. The
dollar-weighted return for TDFs was 5.16% compared with the all funds average of 4.35%.
When comparing dollar-weighted returns of various vintages, as might be expected because
of their heavier equity weightings, later vintages outperformed the nearer vintages. Now that
some actual return data are available, it will be interesting to evaluate performance on a
risk-adjusted basis with actual returns rather than relying on simulations. A recommendation
in this regard is made in the next section.

5. Recommendations

This section is divided into three parts, each aimed at a different set of TDF stakeholders.
The first set of recommendations is directed towards investors and their financial advisors.
The second is for plan sponsors and mutual fund companies and finally and the last set is
aimed at researchers and presented as ideas for future research.

5.1. Investors and financial advisors

In light of the wide variety of TDF offerings, deliberated choices by individuals and their
advisors specific to the individual situation can improve investment outcomes. As with other
mutual fund recommendations, fees are a significant variable affecting long-term results and
the investor or their advisor should closely analyze these. Lipton and Kish (2011) report
front-end loads ranging from 4.75% to 5.75%. They also report that management fees and

255S. Singh / Financial Services Review 25 (2016) 235–262



expense ratios decline with the approach of the target date. Given due diligence and fiduciary
duty requirements, it is important that advisors monitor fund fees over time. Economies of
scale dictate that as assets under management of a fund increase, the expense ratio of the fund
should decline. Investors should avoid a fund that increases fees under such circumstances.

TDFs at a basic level are fund of funds. The underlying funds have their own fees. Many
fund families do not charge overlay fees at the TDF level and the investor pays only the fees
of the underlying funds. Such funds should be preferred to ones that tack on another layer
of fees. The total fees paid by the investor over time should be the ultimate consideration
though.

Passive funds have attracted significant amounts of investor funds and indexing as an
investing strategy is gaining increasing popularity. The empirical evidence against the
long-term performance of actively managed funds and their cost disadvantage do not make
them a competitive alternative in a TDF vis-à-vis those with indexed underlying funds.

There is growing theoretical evidence that a one-size fits all TDF is not an optimal choice.
Incorporating risk tolerance of the investor in the TDF choice provides higher welfare. For
example, an investor whose human capital return has low correlation to equity market return
has perhaps higher risk tolerance in their TDF choices. Under such circumstances, investors
can consider a TDF whose target date is past the intended retirement year. Additionally, a choice
that incorporates increased longevity and longer life spans will likely result in higher utility.

There are two sets of cash flows that are often ignored in the choice and design of TDFs.
These are: (1) The investor’s holdings of real estate in the form of primary residence and (2)
defined benefit pension cash flows like social security. The latter is quite like a real asset with
“bond like” characteristics. According to Jennings and Reichenstein (2003) “the financial
profession ignores the value of the DB plan in calculating asset allocations, it places an
implicit value of zero on DB benefits (p. 197).” Incorporating both of these often-ignored
assets in the TDF choice might indicate a different level of risk aversion and, thus, suggest
a different asset allocation Pederson (2015).

The importance of including real assets in TDFs is emphasized in many studies. In LDI,
many scholars have modeled the retirement liability as a set of real outflows analogous to the
payout on a real annuity. Therefore, inflation protection appears to be a primary objective of
any sound glide path design. Investors and their advisors should favor TDFs that have a
healthy dosage of real assets like TIPS, commodities and other such assets, especially
through retirement. The task of finding utility maximizing choices is facilitated when plan
sponsors and mutual fund companies offer TDFs designed with this goal in mind.

5.2. Plan sponsors and mutual fund companies

Benz (2015) characterizes TDFs as “blunt instruments.” There is widespread agreement
among academics and practitioners that TDF efficiency can be improved by broadly tailoring
funds to investor risk preferences. Here are some simple suggestions for improving effi-
ciency and there are long-term structural changes that can be affected to improve outcomes
for plan participants.

Anecdotally, numerous defined contribution plans offer TDFs from one provider only.
This virtually locks the participant into a glide path that may or may not be relevant to their

256 S. Singh / Financial Services Review 25 (2016) 235–262



risk tolerance and overall portfolio objectives. Plan sponsors can easily increase the menu
choices by offering funds from more than one family. In adding another family of TDFs, it
would be ideal to have new ones that have significantly different glide paths from the current
choice. While this would increase fiduciary responsibility related to monitoring, it would
benefit the sponsor with regard to the fiduciary duty related to “risk appropriateness.” It is
acknowledged that the question of picking the fund best suited to the risk aversion function
of an employee partially defeats the notion of a single Qualified Default Investment Alter-
native (QDIA). However, if this selection is made part of an employee’s onboarding process,
it will lead to more desirable welfare outcomes. For a good exposition on fiduciary respon-
sibility related to TDFs for pension plan sponsors see Landsberg (2014).

The risk profile of the TDF is significantly influenced by the correlation of human capital
returns to equity returns (Bodie and Treussard, 2007; Idzorek, 2008; Shiller, 2005). Bernard
(2009) points out that “target-date funds with a retirement date of 2010 have stock allocations
from 9.15% to 65% of their portfolios.” Prospective buyer of TDFs would appear to have a
wide range of risk choices with which to tailor their holdings. Plan sponsors should strive
to consider work force risk characteristics when offering TDFs to employees. One simple
suggestion is to take into account the correlation of employer stock returns to the broader stock
market. For example, where returns are highly correlated to stock market returns, in line with
recommendations of Bodie and Treussard (2007), a safer TDF might be appropriate.

Stempien and Zoll (2015) report a growing trend towards custom glide paths by plan
sponsors with a unique work force. If the workforce has a similar demographic, for example
85% of the workers are between the ages of 25 to 35, then a custom TDF will have a greater
fiduciary duty demonstrability than an off the shelf product. The advice in this regard from
Bauer, Phillips, and White (2009) is twofold: custom offerings with indexed funds as
underlying options. The benefit of indexed offerings relate to compliance and due diligence
issues while the second of custom TDF address transparency and relevance for the work
force requirements.

With individualized asset allocation through robo-advisors becoming increasingly likely,
it might be possible soon to move from custom target date to individualized TDFs. It may
be possible to develop individualized glide paths for participants or at least provide a
recommendation based on information provided by an employee. The first step in this regard
would be to incorporate risk tolerance information as part of personnel information.

With respect to mutual fund families, it might be worthwhile to offer three variants of the
same target year, or at the very least two. The glide path can be aggressive, balanced, and
conservative or just aggressive and conservative. This will enable investors to tailor risk to
some extent without borrowing or short selling in other parts of the portfolio. Alternatively,
as suggested earlier, this would mitigate the need to seek a TDF of a vintage different from
the intended retirement year.

It is widely acknowledged in the mutual fund industry that Vanguard funds have one of
the lower expense ratios in the industry. In Bell v. Anthem the plaintiff alleges excess fees
charged by the Anthem 401(k) plan that has about $5 billion in assets. All the funds in the
lineup are Vanguard funds. The litigation highlights the need for establishing due diligence
procedures. Tramell (2009) points to the need for monitoring and controlling fund fees as
these can significantly affect retirement outcomes for participants.

257S. Singh / Financial Services Review 25 (2016) 235–262



5.3. Researchers

Given longevity increases and long retirement horizons, an answer to the “to and through
retirement” glide path question is quite vital. This is premised on the basic issue of the
responsibility of the employer or the mutual fund to the employee. Does it end at retirement
or at end-of-life? At this time there seems to be no definitive answer. Each set of providers
claim their glide path to be superior. Researchers should try to answer both aspects of this
question definitively.

The discussion on glide paths earlier in this paper, create a new set of research possibil-
ities. There have been numerous suggestions on an optimal glide path. Blanchett and Straehl
(2015) present a unified single-period framework for incorporating non-tradeable assets like
residential home and human capital in the portfolio optimization process. Their analysis
provides many insights into the design of next generation glide paths and TDFs. The question
of the optimality of a glide path given a utility function that incorporates non-tradeable assets
needs to be explored in greater detail and definitively answered.

Two recent developments in the TDF market have created a host of new research
questions; one set is regulatory and the other technological. The first is the new Treasury rule
that permits the holder of an IRA or 401(K) to invest up to the larger of 25% or $125,000
of the portfolio in a longevity annuity. More important, through Notice 2014–66, the U.S.
Treasury (2014) further permits TDFs to offer deferred annuities as a part of the asset
allocation, without violating discrimination rules, as long as certain conditions are met. One
of the questions examined by researchers previously (Pfau, 2014; Spitzer and Singh, 2008),
focuses on the longevity risk faced with different glide paths during retirement. The inclusion
of longevity and deferred annuities alters the optimal asset allocation profiles of TDFs, and
the actuarial probabilities of longevity risk. A host of new questions need to now be
answered. As the permission to offer annuities is fairly recent, research needs to be
conducted on the optimal pricing of these annuities and their influence on the risk profiles of
TDFs.

The second development is the mechanization of investment advice. The infiltration of
robo-advisors in the investment advice market provides a fertile ground for the implemen-
tation of the idea of individualized glide paths. TDF glide paths that incorporate risk
tolerance individually may not be a far-fetched idea. The day may not be far off where an
individualized glide path is generated for participants based on the information collected
from their personnel file or the initial employment interview.

TDF evaluation mechanisms are still in their infancy. There are a number of indexes now
available to benchmark TDF performance. Each index has strengths and weaknesses. A com-
prehensive comparative index evaluation would provide useful information to defined contribu-
tion plan providers and consultants. Moreover, the impact of building indexes that has underlying
indexes instead of individual securities as components, needs attention and analysis.

One of the other growing and popular approaches to designing glide paths is liability
driven investing or LDI. Currently most LDI related to TDFs assume the retirement liability
to be a set of real cash outflows. To the best of this author’s knowledge, no effort to model
stochastic shocks to accumulation or decumulation cash flows, like unanticipated health care
costs, has been attempted. In the modeling of glide paths incorporating such situations, an

258 S. Singh / Financial Services Review 25 (2016) 235–262



interesting line of inquiry would involve a comparative evaluation of risk tolerance of a
group of employees as opposed to the risk capacity needed to meet their retirement liabilities.
An effort in this regard, would be a valuable addition to the body of knowledge.

6. Conclusion

Aided by a favorable regulatory environment, TDFs have experienced unprecedented flow
of funds over the past eight years. There seem to be two complimentary approaches to the
design of TDF glide path. The first is in the theory of portfolio choice in the presence of
nontraded assets like human capital and the second is rooted in the ultimate goal of meeting
retirement liabilities. Most of the empirical evidence on TDFs calls for better design of glide
paths. Given the long-term horizon of TDF strategies, performance measurement and
evaluation remain a challenge even though indexes are now available to benchmark perfor-
mance. Given the nature of TDFs, none of these indexes fully meets ideal benchmark
qualities. The empirical evidence on, and the design of, TDFs is still in its infancy. Much
work still remains.

Notes

1 During the period 1995 to 2006, TDF assets grew by about $300 billion. Since the
passing of the PPA Act, over the 2007 to 2014 period assets in TDFs increased by
about $700 billion. Bauer, Phillips, and White (2009) predict that TDF assets will
increase by another $1.2 trillion by end of 2018.

2 The author would like to thank an anonymous referee for pointing this fact and for
suggesting an attribution framework.

Acknowledgment

The author would like to thank John J. Spitzer for his review and comments on the article.
The author would also like to thank the editor and three anonymous referees for their
comments and suggestions. Remaining errors and omissions are mine.

References

Agnew, J., Balduzzi, P., & Sundén, A. (2003). Portfolio choice and trading in a large 401(k) plan. American
Economic Review, 93, 193–221.

Arnott, R. D., Sherrerd, K. F., & Wu, L. (2013). The glidepath illusion …. and potential solutions. Journal of
Retirement, 1, 13–28.

Bailey, J. V., Richards T. M., & Tierney, D. E. (2007). Evaluating portfolio performance. In John L. Maginn,
Donald L. Tuttle, Jerry E. Pinto, & Dennis W. McLeavey (Eds.), Managing Investment Portfolios: A Dynamic
Process (pp. 717–782). Hoboken, NJ: John Wiley & Sons, Inc.

259S. Singh / Financial Services Review 25 (2016) 235–262



Balduzzi, P., & Reuter J. (2012). Heterogeneity in target-date funds: Optimal risk taking or risk matching? NBER
Working Paper 17886, Boston, MA.

Bare, R., & Greves, J. (2013). Evaluating target date investment performance. Veiwpoint Russell Research,
Russell Investments Monograph, Russell Investments, Tacoma, WA.

Basu, A. K., & Drew M. E. (2009). Portfolio size effect in retirement accounts: What does it imply for lifecycle
asset allocation funds? Journal of Portfolio Management, 35, 61–72.

Basu, A. K., Byrne, A., & Drew, M. E. (2011). Dynamic lifecycle strategies for target date retirement funds,
Journal of Portfolio Management, 37, 83–96.

Bauer, D. J., Phillips, B. F., & White, J. R. (2009). Target-date retirement funds: The new defined contribution
battleground. Casey Quirk and Associates (available at http://tinyurl.com/p43ppgs).

Benz, C., & Stipp, J. (2015). Morningstar Inc. video, What the Best ‘In-Retirement’ Mutual Funds Can Teach You
(available at http://tinyurl.com/ol647).

Bernard, T. S. (2009). Target-date funds: Seven questions to ask before jumping in. New York Times, June 29
(available at http://tinyurl.com/m9lqdr).

Blanchett, D. M., & Kasten, G. W. (2010). Improving the “target” in target-date investing. Journal of Pension
Benefits, 18, 17–27.

Blanchett, D. M. (2015). Initial conditions and optimal retirement glide paths. Journal of Financial Planning, 28,
52–60.

Blanchett, D. M., & Straehl, P. U. (2015). No portfolio is an island. Financial Analysts Journal, 71, 15–23.
Bodie, Z., & Treussard, J. (2007). Making investment choices as simple as possible, but not simpler. Financial

Analysts Journal, 63, 42–47.
Boscaljon, B. (2011). Determining the “glide path” for target date funds. Financial Services Review, 20, 113–128.
Burtless, G. (2010). Do workers prepare rationally for retirement? In A. Drolet, N. Schwarz, & C. Yoon (Eds.), The

Aging Consumer: Perspectives from Psychology and Economics (pp. 103–127). New York: Taylor & Francis.
Callan Associates (2015). Target date fund index primer (available at http://tinyurl.com/p6kwpwz).
Cassidy, D. P., Peskin M. W., Seigel, L. B., & Sexauer, S. (2014). Be kind to your retirement plan – give it a

benchmark. Journal of Portfolio Management, Special Issue, 33–36.
Chang, R. P., Hunter, D., Qianqiu L., & Saar H. (2014). Saving for retirement while having more nights with

peaceful sleep: Comparison of lifecycle and lifestyle strategies from expected utility perspective, Financial
Services Review, 23, 169–188.

Clark, J. M., & Hood, M. (2009). Policy implications for modeling next generation of target date funds. The
Journal of Investing, 18, 53–61.

Cocco, J. F., Gomes, F. J., & Maenhout, P. J. (2005). Consumption and portfolio choice over the life cycle. The
Review of Financial Studies, 18, 491–533.

D’Antona, J. Jr. (2008). Disappointing performance hits 2010 target-date funds. Pensions & Investments
(available at http://tinyurl.com/mz2l2z).

Dimensional Fund Advisors. (2016). Target date retirement income fund (available at http://tinyurl.com/hgxk8sa).
Dolvin S. D., Templeton, W. K., & Rieber, W. J. (2010). Asset allocation for retirement: Simple heuristics and

target-date funds. Journal of Financial Planning, 23, 60–71.
Employee Benefit Security Administration. (February 2013). U.S. Department of Labor, Washington, DC. Target

Date Retirement Funds—Tips for ERISA Plan Fiduciaries. p. 2–3.
Employee Benefits Security Administration. (April 2008). U.S. Department of Labor, Washington, DC. Regu-

lation Relating to Qualified Default Investment Alternatives in Participant-Directed Individual Account Plans.
p. 1–2.

Estrada, J. (2014). The glidepath illusion: An international perspective. Journal of Portfolio Management, 40,
52–64.

Gomes, F. J., Kotlikoff, L. J., & Viceira, L. M. (2008). Optimal life-cycle investing with flexible labor supply:
A welfare analysis of life-cycle funds. American Economic Review: Papers and Proceedings, 98, 297–303.

Holt, J., Yang, J., Acheson L., & Wasserman, S. (2016). 2016 target-date fund landscape. Morningstar Research,
68.

260 S. Singh / Financial Services Review 25 (2016) 235–262



Idzorek, T., Stempien, J., & Voris, N. (2013). Bait and switch: Glide path instability. Journal of Investing, 22,
74–82.

Idzorek, T. (2008). Lifetime asset allocation methodologies for target date maturity funds. Unpublished manu-
script, Ibbotson Associates, 22–26.

Investment Company Institute. (2015). Investment Company Fact Book, 2015. p. 45, p. 230–231.
Investment Company Institute. (2015). Frequently Asked Questions about Target Date or Lifecycle Funds. p. 2–3:

Investment Company Institute, Washington, DC.
Investment Company Institute. (2014). 401(k) plan asset allocation, account balances and loan activity in 2013,

ICI Research Perspective, 20, 40–47.
Jennings W. W., & Reichenstein, W. (2003). Valuing defined-benefit plans. Financial Services Review, 12,

179–199.
Kalman, M. B. (2011). The role of the equity risk premium in the shortfall risk of target-date funds. Financial

Services Review, 20, 265–284.
Kitces, M. E., & Pfau, W. D. (2015). Retirement risk, rising equity glide paths, and valuation-based asset

allocation. Journal of Financial Planning, 28, 38–48.
Landsberg, R. D. (2014). Fiduciary risk management and target date funds. Journal of Deferred Compensation,

19, 86–94.
Landsberg, R. D. (2011). Target date funds and the proposed SEC rule on disclosure. Journal of Pension Planning

& Compliance, 36, 86–94.
Landsberg, R. D. (2007). An overview of the pension protection act of 2006. Journal of Deferred Compensation,

13, 13–33.
Lewis, N. D. (2008). Assessing shortfall risk in life-cycle investment funds. Journal of Wealth Management, 11,

15–19.
Lipton, A. F., & Kish, R. J. (2011). Life cycle funds: Lack of disclosure and lack of return. Financial Services

Review, 20, 75–94.
Markowitz, H. M. (1952). Portfolio selection. Journal of Finance, 7, 1, 77–91.
Mayers, D. (1972). Non-marketable assets and capital market equilibrium under uncertainty. In M. Jensen (Ed.),

Studies in the theory of capital markets (pp. 223–248). New York: Praeger.
McGraw Hill Financial. (2015). Dow Jones target date indexes: Risk attributes and performance (available at

http://tinyurl.com/nsqk3py).
Meder, A. H. (2012). Level 1 LDI: Selecting an appropriate benchmark. Journal of Investing, 21, 117–127.
Merton, R. C. (1969). Lifetime portfolio selection under uncertainty: The continuous-time case. Review of

Economics and Statistics, 51, 247–257.
Morningstar Inc. (2015). Target date indexes: Comparison grid (available at http://tinyurl.com/o42pbdf).
Nagenast, J. C., Bucci, J., & Coaker, W. J., II. (2006). Popping the hood: An analysis of major life cycle fund

families (2nd ed.). Turnstone Advisory Group LLC Marina Del Ray, CA.
Pederson, N. (2015). Rethinking retirement risk (available at http://tinyurl.com/ngj227c).
Pfau, W. D. (2010). Lifecycle funds and wealth accumulation for retirement: Evidence for a more conservative

asset allocation as retirement approaches. Financial Services Review, 19, 59–74.
Pfau, W. D. (2011). An optimizing framework for the glide paths of life cycle asset allocation funds. Applied

Economic Letters, 18, 55–58.
Pfau, W. D., & Kitces, M. E. (2014). Reducing retirement risk with a rising equity glide path. Journal of

Financial Planning, 27, 38–45.
Richardson, D. P., & Bissette, B. (2014). Trends in premium and asset allocation by TIAA-CREF participants:

2005–2011. TIAA-CREF Institute Research Dialogue, 112, 23–24.
Sharpe, W. F. (1964). Capital asset prices: A theory of market equilibrium. Journal of Finance, 19, 3, 425–442.
Shiller, R. J. (2005). Life-cycle portfolios as government policy. The Economists’ Voice, 2, 1–14.
Spitzer J. J., & Singh S. (2011). Assessing the effectiveness of lifecycle (target-date) funds during the accumu-

lation phase. Financial Services Review, 20, 327–341.
Spitzer, J. J., & Singh, S. (2008). Shortfall risk of target-date funds during retirement. Financial Services Review,

17, 143–153.

261S. Singh / Financial Services Review 25 (2016) 235–262



Stempien, J., & Zoll A. (2015). Morningstar Inc. video, What’s Behind Target-Date Funds’ Popularity?
(available at http://tinyurl.com/owznmlz).

Surz, R. J., & Israelsen, C. L. (2007). Evaluating target date lifecycle funds. The Journal of Performance
Measurement, Winter, 62–70.

Tang, L., & Lin Y.-T. (2015). The efficiency of target-date funds. Journal of Asset Management, 16, 131–148.
Tang, N., & Mitchell, O. (2010). The efficiency of sponsor and participant portfolio choices in 401(k) Plans.

Journal of Public Economics, 94, 1073–1085.
Torregrosa, P. T., & Koenig, D. A. (2008). Target date funds: The case for a more conservative glide path.

Institutional Investor Guide to Defined Contribution and Defined Benefit Services, Wells Fargo Bank.
Towers Watson. (2010). Defined contribution: Lessons after the crisis (available at http://preview.tinyurl.com/

nqorkdp).
Tramell, S. (2009). Glide path. CFA Magazine, Jan-Feb 2009, 36–43.
United States Treasury. (2014). Treasury issues guidance to encourage annuities in 401(k) plans (available at

http://tinyurl.com/jr8nc82).
Vanguard Group. (2015). Vanguard target retirement funds prospectus. January 27, 62x.
Viceira, L. M. (2009). Life cycle funds. In A. Lusardi (Ed.), Overcoming the saving slump: How to increase the

effectiveness of financial education and saving programs (pp. 140–178). Chicago, IL: University of Chicago
Press.

Whitton, B., & Thuerbach, K. (2015). PIMCO’s glide path: An outcome focused design. PIMCO Featured
Solution, March, 2–8.

Williams, J. (1978). Risk, human capital, and the investor’s portfolio. Journal of Business, 51, 65–89.
Yoon, Y. (2010). Glide path and dynamic asset allocation of target date funds. Journal of Asset Management, 11,

346–360.

262 S. Singh / Financial Services Review 25 (2016) 235–262