Saving for Retirement: The Effects of Fund Assortment Size and Investor Knowledge on Asset Allocation Strategies

Authors

  • MAUREEN MORRIN,

    1. 1 Maureen Morrin is an associate professor of marketing in School of Business, Rutgers University, Camden, NJ (mmorrin@Rutgers.edu). 2Susan Broniarczyk is a professor in the Department of Marketing, McCombs School of Business, University of Texas at Austin (susan.broniarczyk@mccombs.utexas.edu). 3J. Jeffrey Inman is Albert Wesley Frey Professor of Marketing and a professor of business administration at the Joseph M. Katz Graduate School of Business, University of Pittsburgh (jinman@katz.pitt.edu). 4John Paul Broussard is an associate professor of finance at the School of Business, Rutgers University, Camden, NJ (broussar@camden.rutgers.edu).
    Search for more papers by this author
  • 1 SUSAN BRONIARCZYK,

    1. 1 Maureen Morrin is an associate professor of marketing in School of Business, Rutgers University, Camden, NJ (mmorrin@Rutgers.edu). 2Susan Broniarczyk is a professor in the Department of Marketing, McCombs School of Business, University of Texas at Austin (susan.broniarczyk@mccombs.utexas.edu). 3J. Jeffrey Inman is Albert Wesley Frey Professor of Marketing and a professor of business administration at the Joseph M. Katz Graduate School of Business, University of Pittsburgh (jinman@katz.pitt.edu). 4John Paul Broussard is an associate professor of finance at the School of Business, Rutgers University, Camden, NJ (broussar@camden.rutgers.edu).
    Search for more papers by this author
  • 2 J. JEFFREY INMAN,

    1. 1 Maureen Morrin is an associate professor of marketing in School of Business, Rutgers University, Camden, NJ (mmorrin@Rutgers.edu). 2Susan Broniarczyk is a professor in the Department of Marketing, McCombs School of Business, University of Texas at Austin (susan.broniarczyk@mccombs.utexas.edu). 3J. Jeffrey Inman is Albert Wesley Frey Professor of Marketing and a professor of business administration at the Joseph M. Katz Graduate School of Business, University of Pittsburgh (jinman@katz.pitt.edu). 4John Paul Broussard is an associate professor of finance at the School of Business, Rutgers University, Camden, NJ (broussar@camden.rutgers.edu).
    Search for more papers by this author
  • and 3 JOHN BROUSSARD 4

    1. 1 Maureen Morrin is an associate professor of marketing in School of Business, Rutgers University, Camden, NJ (mmorrin@Rutgers.edu). 2Susan Broniarczyk is a professor in the Department of Marketing, McCombs School of Business, University of Texas at Austin (susan.broniarczyk@mccombs.utexas.edu). 3J. Jeffrey Inman is Albert Wesley Frey Professor of Marketing and a professor of business administration at the Joseph M. Katz Graduate School of Business, University of Pittsburgh (jinman@katz.pitt.edu). 4John Paul Broussard is an associate professor of finance at the School of Business, Rutgers University, Camden, NJ (broussar@camden.rutgers.edu).
    Search for more papers by this author

  • The authors gratefully acknowledge generous financial support for this project from a grant by the FINRA (formerly NASD) Investor Education Foundation (#2005-080).

Abstract

We report the results of a decision simulation conducted among 211 adults whose task was to invest in a 401(k) retirement plan. We varied the number of mutual funds (three vs. twenty-one) offered for investment and assessed investing knowledge with a self-report measure. The results indicate that less knowledgeable (but not more knowledgeable) investors change their asset allocation strategies as a function of fund assortment size, such that a significantly higher proportion of dollars invested by the less knowledgeable investors is allocated to stocks when choosing from the larger assortment.

The retirement investment decisions faced by employees are becoming more daunting as employers increase the number of mutual funds they offer in their defined contribution plans. The Vanguard Center for Retirement Research notes that the average number of funds offered in the plans they manage has risen from twelve to fifteen in just four years, with more than 60% of their plans offering more than ten options (http://www.403bwise.com/pdf/vcrr_choice.pdf). What impact do these larger fund assortments in 401(k) plans have on retirement investors’ decisions? Do larger assortments result in systematic shifts in investor behavior? If such effects are evident, are they more pronounced for less knowledgeable investors? These are the key issues explored in the present research.

Compared to less knowledgeable investors, those possessing more knowledge are more likely to be aware of the importance of asset allocation to a portfolio’s long-term performance and are more likely to incorporate allocation strategies into their decision strategies. The notion that asset allocation helps reduce risk for a given level of return and thus enhances investment decision quality is based on modern portfolio theory of Markowitz (1952). He suggested that investments should not be viewed in isolation but rather in combination with other investments. Research in the finance literature suggests that portfolio diversification, rather than the choice of individual investments within an asset class or attempts at market timing, accounts for the vast majority of long-term investment performance. Indeed, Samuelson (1990) argues that market-timing attempts in terms of asset allocation changes can be very detrimental to portfolio diversification.

Brinson, Hood, and Beebower (1986; Brinson, Singer, and Beebower 1991) examined 91 large U.S. pension funds from 1974 to 1983 and found that 93.6% of the variation in returns were accounted for simply by how assets were allocated across the three investment classes of stocks, bonds, and cash equivalents (such as money market funds), rather than other variables such as specific investments chosen or market timing (see also Bernstein 2000; Hooks 1998). This finding has been found to be true not only when investing in individual stocks but also when investing in mutual funds (Niendorf and Lang 1995; Sharpe 1992) and when investing outside the United States (Ferruz, Vincente, and Andreu 2007). Thus, in this research, we focus on fund assortment effects on asset allocation decisions.

Below, we review the literatures on product assortment and knowledge and develop our hypothesis. We then report the results of a study conducted among 211 adult consumers, which examined the effects of fund assortment size on investor decision making in a simulated 401(k) plan choice setting. We conclude with policy implications and ideas for future research.

Assortment

Research suggests that consumers prefer choosing from larger assortments (Broniarczyk 2008; Kahn and Lehmann 1991). Larger choice sets should increase the likelihood that a consumer is able to choose a product or a service that best meets his or her needs (Baumol and Ide 1956). A significant literature has arisen that shows that product assortments can significantly influence both what is preferred and what is purchased by consumers (for reviews, see Broniarczyk 2008 and Simonson 1999).

However, some of the negative psychological effects of choosing from larger product assortments have gained attention in the past few years (Botti and Iyengar 2006; Schwartz 2004). Researchers have found that large assortments can create confusion (Huffman and Kahn 1998; Morales et al. 2005) and information overload for consumers (Gourville and Soman 2005), some of whom delay their choice (Greenleaf and Lehmann 1995), or simply decide not to make a decision, and walk away from the choice task at hand (Chernev 2005; Gourville and Soman 2005; Iyengar and Lepper 2000). Huberman, Iyengar, and Jiang (2003) investigated the impact of fund assortment size on consumers’ participation in retirement plans. They found that every ten funds added to a mutual fund assortment led to a 1.5%–2.0% drop in participation rates.

More recently, research has shown how the impact of product assortment on individuals’ decisions is moderated by the degree to which they have articulated preferences, that is, the degree to which a decision maker possesses a “readily available ideal combination of various product attributes” (Chernev 2003, p. 152). Those with articulated preferences are more likely to search for a choice or set of choices that meets their desired set of attributes. Those without such ideal points readily available do not have such a reference point for evaluating choice alternatives. This line of research suggests that there may exist individual difference variables that moderate the effects of larger product assortments on choice outcomes.

The current project examines whether larger mutual fund assortments lead to systematic shifts in individuals’ retirement investment strategies as a function of investor knowledge. Here, we explore the impact of fund assortment size on the proportion of dollars invested in stocks, bonds, and cash (i.e., money market). We are interested in how investment strategies shift if one is offered a plan that contains three funds from which to choose versus a plan than contains twenty-one funds from which to choose while controlling for proportion of funds offered within each major asset class. The key moderating variable of interest is investor knowledge.

Knowledge

Research has examined how experts (vs. novices) acquire, store, and use information in the decision process (e.g., Alba and Hutchinson 1987; Brucks 1985; Johnson and Russo 1984; Mitchell and Dacin 1996). This stream of research has shown that experts, who are generally more knowledgeable in the relevant decision-making domain, develop more refined cognitive structures that they use to differentiate among product alternatives within that domain (Alba and Hutchinson 1987). Thus, individuals with more knowledge about investing should be able to more easily and more rapidly differentiate among classes of mutual funds, such as those that invest in stocks, bonds, or cash. More knowledgeable investors are more likely to have articulated preferences as well and thus will be more able to compare the available choices to an internal reference point.

More knowledgeable investors are more likely to process more abstract-level knowledge, including strategies regarding asset allocation, and are thus more likely to use asset allocation goals (e.g., to invest X% in stocks and Y% in bonds). Those with greater levels of investing knowledge are also more likely to better understand the theory behind and benefits of portfolio diversification. Those with less knowledge should have more difficulty categorizing and differentiating funds along meaningful attributes such as by asset class. Instead, being less sure of their ideal point or desired portfolio attributes, they may be more likely to be impacted by structural aspects of the plan, such as the sheer number and variety of funds offered for investment.

Based on the preceding, we expect that more knowledgeable investors’ asset allocation strategies will not be significantly impacted by fund assortment structure. However, less knowledgeable investors’ asset allocations strategies will be. Alba and Hutchinson (1987) suggested that problem solving by less knowledgeable decision makers is more influenced by external factors such as point-of-purchase displays. Retirement 401(k) plan structure is an analogous external cue to which less knowledgeable investors should be more vulnerable. For example, those with little knowledge might perceive greater variety associated with a larger stock fund assortment (e.g., greater variance in returns) and respond to the greater perceived increase in variety with increased consumption of that asset class. Prior research has shown that offering greater variety or even just greater perceived variety in product assortments increases the amount consumed (Kahn and Wansink 2004). This leads to our key hypothesis, which is that less but not more knowledgeable investors choosing from a larger (vs. smaller) fund assortment size will invest a significantly larger proportion of their dollars in stocks (vs. bonds or cash). We describe below a study designed to test this hypothesis.

Method

Design

The study we conducted consisted of a single-factor (fund assortment size: small or large) between-subjects design. Assortment size was manipulated such that the plan offered either three or twenty-one mutual funds for investment (i.e., a small or large fund assortment; Appendix 1, the funds are listed in alphabetical order). In the small assortment condition, each of the three asset classes offered a single fund for investment (one stock fund, one bond fund, and one money market fund). In the large assortment condition, each of the three asset classes offered seven funds for investment (seven stock funds, seven bond funds, and seven money market funds). Thus, across both conditions, participants were offered equal proportions of stock, bond, and money market funds from which to choose.

Sample

We conducted the experiment among a convenience sample of adults (N = 211, 58% male, 77% employed full time; Table 1). Participants were selected by asking graduate students enrolled in business classes to enlist up to three adults they knew to fill out a questionnaire for class credit. The respondents were entered into drawings for $100 lotteries as a token of appreciation. The majority of respondents worked full time and were not students. Respondents were randomly assigned to one of the two cells of the study (i.e., small or large fund assortment condition). Since other research has examined participation rates as a function of fund assortment size (e.g., Huberman, Iyengar, and Jiang 2003), our focus in this study was on asset allocation, conditional on plan participation. The students were tasked to obtain completed surveys and all respondents chose to participate in the plan.

Table 1. 
Demographic Characteristics of the Respondent Sample (N= 211)
CharacteristicPercentage of Sample
Gender
 Female42
 Male58
Native language
 English91
 Other9
Age (in years)
 Younger than twenty2
 Twenty38
 Thirty22
 Forty21
 Fifty12
 Sixty4
 Seventy1
Have a 401(k) plan
 No19
 Yes81
Have a pension plan
 No35
 Yes65
Employment status
 Full time77
 Part time12
 Student9
 Unemployed/retired2
Household income
  $0–$25,0009
  $25,001–$50,00016
  $50,001–$75,00021
  $75,001–$100,00021
 More than $100,00034

Procedure

Participants were randomly assigned to the small or large assortment condition via a random distribution of booklets to the graduate students responsible for obtaining respondents. Each participant received a booklet that asked them to imagine that they were an employee of a firm that offered them the opportunity to invest in a 401(k) plan. Participants were exposed to information describing the plan, including descriptions about each of the funds offered for investment. Participants then decided how much money to invest and how they would invest their monies across the mutual fund options provided. They were informed about the maximum annual dollar investment they could make and that this maximum was set by the federal government.

The fund descriptions were based on actual funds available at Vanguard, although Vanguard-specific, brand-identifying information was removed from the stimuli. After the investment decision was completed, respondents answered several questions on 5-point Likert items (from 1 = Strongly Disagree to 5 = Strongly Agree). One of these items asked the extent to which they agreed with the following statement: “Compared to most people, I know a lot about investing.” Respondents’ answers to this question were used as a self-report measure of investing knowledge. Although multiple-item measures are often preferable when measuring complex constructs, recent research indicates that the predictive validity of single-item measures for concrete constructs are as predictive as multi-item measures (Bergkvist and Rossiter 2007). In the analyses reported below, we used responses on this item as a continuous variable for the regression analyses and conducted a median split on this item for the analyses of variance (ANOVAs). Respondents also answered manipulation check and demographic questions. Unless otherwise noted, the analyses reported below are based on regressions, which are followed up with two-way ANOVAs for mean comparisons.

Results

Manipulation check

Toward the end of the questionnaire, respondents were asked whether, to the best of their knowledge, the following statement was true or false: “This firm offers a wide variety of funds for investment.” We ran a logistic regression on responses to this item as a function of fund assortment size (small or large), investor knowledge, and their interaction. The only significant effect was fund assortment size, Wald(1) = 16.77, p < .0001. As expected, a higher proportion of respondents in the large assortment condition agreed to this statement compared to those in the small assortment condition (88% vs. 23%, respectively).

Percent invested in stocks

We ran a regression on percentage of portfolio invested in stock funds as a function of fund assortment size (small or large), investor knowledge, and their interaction, plus the demographic covariates of age, gender, and household income. The model was significant, F(6, 177) = 9.79, p < .0001, R2= .25, with both the main effects and the interaction statistically significant. Fund assortment size had a positive impact on percent invested in stocks (t = 4.35, p < .0001, beta = .666; Tables 2 and 3). Knowledge also had a positive impact on percent invested in stocks (t = 5.16, p < .0001, beta = .487). The interaction between fund assortment size and knowledge (t =−2.70, p < .01, beta =−.425) indicated that fund assortment size had a greater effect on those less knowledgeable about investing. The large fund assortment more than doubled investment in stocks among those less knowledgeable, from 28.7% to 60.2% (p < .0001), whereas the increase for those more knowledgeable was not statistically significant (60.7%–65.6%, p > .35). This result supports our hypothesis.

Table 2. 
Mean Results by Conditiona
 Three Fund AssortmentsTwenty-One Fund Assortmentsp
  • a

    Covariate-adjusted means.

% Invested in stock funds
 Low knowledge28.760.2.000
 High knowledge60.765.6.373
% Invested in bond funds
 Low knowledge45.526.2.000
 High knowledge27.521.5.226
% Invested in money market funds
 Low knowledge21.113.5.064
 High knowledge11.811.2.888
Table 3. 
Regression Results
Dependent VariableEffectBSEBetatp
% Invested in stock fundsConstant.077.085 0.906.366
Gender.041.043.0670.971.333
Age −.019.017 −.078 −1.139.256
Income.029.016.1291.796.074
Assortment.404.085.6664.351.000
Knowledge.118.023.4875.158.000
Assortment × knowledge −.087.032 −.425 −2.697.008
% Invested in bond fundsConstant.607.075 8.093.000
Gender −.020.038 −.038 −0.520.604
Age.011.015.0560.783.435
Income −.013.014 −.069 −0.913.363
Assortment −.293.083 −.574 −3.549.000
Knowledge −.080.020 −.391 −3.939.000
Assortment × knowledge.059.029.3452.068.040
% Invested in money market fundsConstant.322.062 5.200.000
Gender −.020.031 −.049 −0.632.528
Age.006.012.0400.528.598
Income −.018.012 −.117 −1.469.144
Assortment −.109.068 −.272 −1.597.112
Knowledge −.038.062 −.234 −2.240.026
Assortment × knowledge.024.024.1751.000.319

Percent invested in bonds

We next conducted a similar regression analysis on percent invested in bonds, to see whether those less knowledgeable were sourcing their increased investment in stocks from this asset class (vs. from cash) when choosing from the large assortment size. The bond fund regression, F(6, 178) = 5.76, p < .0001, R2= .16, produced significant effects for fund assortment size (t =−3.55, p < .0001, beta =−.574), knowledge (t =−3.94, p < .0001, beta =−.391), and their interaction (t = 2.07, p < .05, beta = .345). This regression indicates that the percent invested in bonds declined as fund assortment size increased and as knowledge increased and that the impact of fund assortment size was greater for those less knowledgeable. Specifically, the large fund assortment significantly decreased investment in bonds for those less knowledgeable, from 45.5% to 26.2% (p < .0001), whereas the decrease for those more knowledgeable, from 27.5% to 21.5%, was not statistically significant (p > .22). Thus, those less knowledgeable were indeed shifting their dollars from bonds to stocks when choosing from the larger fund assortment size.

Percent invested in cash

A similar regression was conducted on percent invested in money market funds, F(6, 178) = 2.34, p < .05, R2= .07. The results indicated that only knowledge was a significant determinant of percentage of portfolio invested in money market funds (t =−2.24, p < .05, beta =−.234), with those more knowledgeable investing less in money market funds than those less knowledgeable (regardless of fund assortment size). Thus, the larger fund assortment caused less knowledgeable investors to shift their dollars from bond funds to stocks funds, leaving the proportion allocated to money market funds largely unchanged.

Discussion

The findings of this study support the notion that less (vs. more) knowledgeable investors are more likely to change the asset composition of their portfolios in response to choosing from a larger fund assortment. Specifically, the proportion of dollars allocated to stocks (vs. bonds or cash) more than doubled for less knowledgeable investors, whereas it had no significant impact on the allocation strategies of those more knowledgeable. This result suggests that substantive aspects of less knowledgeable investors’ portfolios may be subject to menu effects or structural characteristics of 401(k) plans. While it is not necessarily undesirable for less knowledgeable investors to allocate more of their dollars to stocks, it is disconcerting that merely changing the total number of funds offered in the plan while controlling for asset class proportions offered in the plan has such a large impact on the risk profile of their investment portfolio.

Future research could focus on better understanding the drivers of this notable effect. We have suggested that it may be due to the effect of perceived variety among the different asset classes. Thus, when offered a wider variety of stock funds from which to choose, as well as an equally wider variety of bond funds from which to choose, it is possible that the subset of stock funds is perceived to offer more variety than the subset of bond funds, among those less knowledgeable about investing. Such an interpretation would not be unlikely, given the greater range in returns typically exhibited by stock versus bond funds. This possibility and others could be tested in future research.

Future research could also explore other potential effects of plan structure on investor behavior. For example, do those with limited knowledge respond differently than do those with more knowledge to fund categorization? For example, would less knowledgeable investors be more likely to invest across a wider variety of funds if they were categorized by investment style (e.g., large cap, small cap, value, growth, international, real estate, health care)? Would less knowledgeable investors be more impacted by whether the mutual funds offered in a 401(k) plan were offered by a single fund family (e.g., Vanguard) versus multiple fund families (e.g., Vanguard, Fidelity, T. Rowe Price, Putnam)? Further research is also warranted to examine whether the subjective knowledge effects examined here generalize to objective knowledge effects on asset allocation strategies (Mitchell and Dacin 1996). Finally, beyond identifying these and other types of menu effects to which less knowledgeable investors might be particularly prone to respond, future research could investigate potential ways to alleviate such effects. One potential solution to some of the issues associated with ever expanding fund assortments is the opportunity to invest in lifecycle or target retirement date funds, which automatically allocate an investor’s dollars by asset class and readjust the allocations as the investor ages.

Investors who do not consider themselves highly knowledgeable about investing are likely to be the greatest beneficiaries of the recent federal legislation designed to encourage employers to offer 401(k) opt-out plans. In these plans, employees are automatically enrolled unless they choose to opt out. This type of default option should significantly increase plan participation, especially among investors with limited knowledge, who often lack the confidence to make the decision to participate in defined contribution plans. Similarly, a default allocation strategy might be beneficial for less knowledgeable investors, such that a traditional allocation (e.g., 60% stocks, 40% bonds) might be the automatic choice for individuals signing up for a 401(k) plan unless they opt to tailor their own investment allocations. In this way, participation in 401(k) plans would no longer be perceived as the daunting task they have been in the past. Moreover, less knowledgeable investors would be much less subject to menu effects such as those demonstrated in the current research. However, it is possible that the less knowledgeable may actually be less likely to take advantage of such default allocation plans, if they perceive greater variety in the traditional choice option or are miscalibrated and overconfident regarding their knowledge level (Alba and Hutchinson 2000). Such a possibility represents an interesting and important area for follow-up research.

Appendix

Table APPENDIX1. 
Small and Large Fund Assortments
Fund NameFund ObjectiveAverage Annual Total Returns (%)
One YearThree YearsTen Years
Small fund assortment (three funds)
 500 Index Stock FundThe fund seeks to track the performance of a benchmark index that measures the investment return of large capitalization stocks.20.6 −10.210.4
 Federal Money Market FundInvests primarily in short-term securities that are issued by U.S. government agencies.1.02.74.4
 Long-Term Corporate Bond FundSeeks current income by investing primarily in high-quality corporate bonds, with an average maturity of fifteen to twenty-five years. The fund’s expense advantage allows it to pursue a higher level of income with less risk than comparable funds.9.811.47.3
Large fund assortment (twenty-one funds)
 500 Index Stock FundThe fund seeks to track the performance of a benchmark index that measures the investment return of large capitalization stocks.20.6 −10.210.4
 Admiral Treasury Money Market FundInvests solely in direct government obligations, such as U.S. Treasury bills and other short-term securities backed by the full faith and credit of the U.S. government. This fund’s expenses are low because of its high minimum investment.1.12.74.3
 Explorer Stock FundSeeks long-term capital growth by investing primarily in the stocks of smaller companies. This fund’s advisers use both fundamental (company, industry, and economic research) and quantitative (computer modeling) analysis to select stocks that have significant growth potential based on the advisers' judgments about companies' financial prospects.41.6 −3.210.8
 Federal Money Market FundInvests primarily in short-term securities that are issued by U.S. government agencies.1.02.74.4
 GNMA Bond FundSeeks current income by investing primarily in Government National Mortgage Association (“Ginnie Mae”) securities, which are backed by the U.S. government to provide timely payment of principal and interest (yield and share price are not guaranteed).2.77.76.8
 Growth Equity Stock FundSeeks long-term capital growth by investing in the stocks of midsize and large companies with strong earnings prospects, and selling those whose earnings prospects are deteriorating. This fund’s adviser evaluates these earnings prospects through a blend of computer-driven and fundamental (company, industry, and economic) analysis.28.6 −23.67.4
 Health Care Stock FundSeeks long-term capital growth by investing in U.S. and foreign companies that develop, produce, or distribute products and services related to health care. These include pharmaceutical firms, medical supply companies, companies that operate health care facilities, and companies engaged in research. To discourage short-term trading, the fund assesses a 1% redemption fee on shares held less than five years.16.00.619.6
 High-Yield Tax-Exempt Bond FundSeeks high current income exempt from federal tax by investing primarily in medium-quality municipal securities, with an average maturity of fifteen to twenty-five years. This fund offers the highest yields but is subject to the highest risk of principal fluctuation.6.67.05.7
 International Growth Stock FundSeeks long-term capital growth by investing in the stocks of foreign companies believed by its investment advisers to exhibit above-average growth potential. To maintain geographic diversity, the fund’s advisers invest in a number of international stock markets; most investments are made in Europe and in the Pacific region. To discourage short-term trading, the fund assesses a 2.0% fee on redemptions of shares purchased on or after June 27, 2003, and held less than two months. The fee is paid directly to the fund and therefore is not considered a load.25.1 −8.25.2
 Intermediate-Term Bond Index FundSeeks to track the performance of a market-weighted bond index with an intermediate-term, dollar-weighted average value.7.510.4
 Long-Term Corporate Bond FundSeeks current income by investing primarily in high-quality corporate bonds, with an average maturity of fifteen to twenty-five years. The fund’s expense advantage allows it to pursue a higher level of income with less risk than comparable funds.9.811.47.3
 NJ Tax-Exempt Money Market FundInvests primarily in high-quality New Jersey municipal money market securities. This fund provides income that is exempt from both federal and New Jersey personal income taxes.0.91.82.7
 PA Tax-Exempt Money Market FundInvests primarily in high-quality Pennsylvania municipal money market securities. This fund provides income that is exempt from both federal and Pennsylvania personal income taxes.0.91.92.9
 Prime Money Market FundInvests in a combination of commercial paper, certificates of deposit, bankers' acceptances, and U.S. government securities. This fund typically offers the highest yield of our money market funds.1.02.74.4
 Short-Term Corporate Bond FundSeeks current income by investing primarily in high-quality corporate bonds, with an average maturity of one to three years. This fund’s expense advantage allows it to pursue a higher level of income with less risk than comparable funds.5.26.65.9
 Short-Term Tax-Exempt Bond FundSeeks current income exempt from federal tax by investing primarily in high-quality municipal securities, with an average maturity of one to two years. This fund pursues a higher level of income than that provided by comparable funds.2.33.83.8
 Short-Term Treasury Bond FundSeeks current income by investing primarily in direct government obligations, such as U.S. Treasury notes and other securities backed by the full faith and credit of the U.S. government, with an average maturity of one to three years. This fund pursues a higher level of income than that provided by comparable funds.2.77.15.8
 Tax-Exempt Money Market FundInvests in high-quality municipal securities issued by state and local governments across the United States. This fund provides income that is exempt from federal tax.1.02.02.9
 Treasury Money Market FundInvests solely in direct government obligations, such as U.S. Treasury bills and other short-term securities backed by the full faith and credit of the U.S. government. This fund offers the highest credit quality available.0.92.64.1
 Value Index Stock FundThe fund seeks to track the performance of a benchmark index that measures the investment return of large capitalization value stocks.23.8 −6.39.6
 Windsor Stock FundSeeks long-term capital growth and current income by investing primarily in the stocks of large and midsize companies believed by the advisers to have superior return potential not reflected in their current prices. This fund’s advisers use both fundamental (company, industry, and economic research) and quantitative (computer modeling) analysis to identify those out-of-favor securities that will outperform the market over time.30.72.210.6

Ancillary