Conlribiilioiis

For tin

Indexing Versus Active Mutual Fund Management

by Rich Fortin, Ph.D., and Stuart Michelson, Ph.D.

The purpose of this paper is to examine the benefits of active mutua! fund management investing versus tndex funds, in general, we find that index funds outperform actively managed funds for most equity and alt bond fund categories on both a total return and after-tax total return basis, with the exception of actively managed Small Company Equity (SCE) and International Stock (tS) funds.These results should be viewed with caution, however, as there is evidence that actively managed funds outperform the index funds during periods when the economy is either going into or out of a recession.

Rich Fortin, Ph.D., is a professor of finance at New Mexico State University in Las Cruces, New Mexico. He has written articles (or the Journal of Financial and Quantitative Analysis; Journal of Financial Research; Journal of Investing; Journal of Financial Planning; Financial Analysts Journal; and the Journal of Small Business Finance.

Stuart Michelson, Ph.D.. is the Roland & Sarah George Professor of Finance at Stetson University in DeLand, Florida. He has written articles for the Journal of Business, Finance, and Accounting; Financial Services Review; Journal of Investing; Journal of Financial Planning; British Accounting Review; and the Journal of Financial Education.

'his paper examines the benefits of active mutual fund managcnienr versus investing in index funds. Do actively managed funds perform \v ell enough to overcome the cost and tax disadvantages? This study is an update and extension of a prior paper by Fortin and Michelson (1999) in that a more recent and longer time period is covered, and comparisons are made across investment categories, and over time, to actual index funds rather than market indexes. An after-tax return comparison is added to the analysis. There has been a longstanding discussion over the relative benefits oi active versus passive management in the mutual fund literature. i)n the one hand, the very fact that we have thousands of investment professionals involved in active mutual fund management suggests that there must he benefits accruing to supposedly rational
Journal of Financial Planning/September 2002

investors in these funds. For example, Elton, (iruber and Blake (1996) show that their portfolio of high-alpha actively managed funds outperformed the Vanguard S&P Index fund from 1981 to 1993, Lowenstein (1997) debates whether indexing is affecting underlying stock prices. He indicates that there may he a premium for stocks that are added to the S&P 500 and that the very nature of indexing creates overvaluation of indexed stocks, Wermers (2000) finds that equity mutual funds outperform the market by 1.3 percent per year, although expenses and transaction costs reduce this benefit to essentially zero. His conclusion: "Funds pick stocks well enough to cover their costs." On tbe other hand, both recent and long-term evidence points to the advantages of indexing over active management, Ftton, Gruber and Blake (1996) ask the relevant question: "Given that there are sufficient index funds to span most investors' risk choices, that the index funds are available at a low cost, and that the low cost of

index funds means that a combination of index funds is likely to outperform an active fund of similar risk.,, why select an actively managed fund?" Bogle (2000) illustrates that an index fund has a 350-b3sis-point advantage over the average equit\' mutual fund due to management expenses, brokerage costs, sales charges and tax advantages. Arnott, Berkin and Ye (2000) find that the Vanguard 500 Index fund outperfonns the average equity mutual fund and the effect is amplified wben taxes are considered, Malkiel (1996) notes that over the past 25 years, about 70 percent of active equity managers have been outperformed by the S&P 500 Stock Index. Gruher (1996) and Bogle (1995) also find similar results. They argue that index funds allow investors to buy securities of many different types with minimal expen.se and significant tax savings. Bogle (1996) states that "the case for selecting an index fund is compelling due to indexing's inherent cost advantage." Malkiel (1995) concludes by stating that

Fortin

Contributions

"most investors would be considerably better off by purchasing a low expense index fund than b}' trying to seleet an active fund manager who appears to possess a hot band." Tbis study tests the hypothesis that actively managed mutual funds have significantly outperformed (imdei'j.ierfomied) index funds over the study period, 1976-2000. We perforni this investigation using both total returns and after-tax total returns, whicb has not been done previously. Tbis sttidy examines eight classes of mutual fund categories, including multiple categories of equit)' funds and bond funds. Previous research primarily examined smaller samples and time periods of equity funds on only a before-tax basis.

Vanguard Small Cap Index, Vanguard Total International Index and Vanguard Balanced Index. Tbese index funds were selected because of their more complete return series over tbe study period and the fact that the Vanguard (iroup pioneered index investing over 25 years ago. The final sample contains 9,329 funds with 46,540 annual return data points.' The active])' managed fund categories were matcbed with index funds that most closely approximated the investment objective of the funds. A consideration in the selection of the comparison index fund was that tbe index fund had a return series long enough to perform statistical tests. Tbe funds were matched as follows. The Aggressive Growth & Growth, Growth/In come &. Fquity/Income and Specialty Equity funds were matched with the Vanguard Index 500 fund. Small Company Equity funds were matched with the Vanguard Small Cap Index fund. The (Corporate Bond and Ciovernment Bond funds w ere matcbed with tbe Vanguard Total Bond Index fund. The International Stock funds w ere matcbed with the \''anguard Total International Index fund. The Asset Allocation & Balanced funds were matehed with tbe \'anguard Balanced Index fund." Results also were computed for tbe Vancuard Total Stock Index fund and Vantjuard Extended Market Index fund, but because the return series for these tunds were much shorter, the results are not presented in this paper.

testing the mean (median) differences in fimd returns between tbe index funds and eaeb categorv'. Botb parametric and nonparametric statistical tests are used wbere applicable. Tests are performed on both a before- and after-tax basis. (A parametric test is a statistical test of significance primarily used for larger samples tbat are normally distributed, A nonparametric test is a statistical test of significance primarily used for smaller samples, v\ ithout the restriction that the sample be normally distributed.) Because annual total returns (calculated assuming reinyestment of all diyidends and capital-gain distributions) are proyided by Morningstar, tbe critical variable to be computed is the after-tax total return. Tbis calculation involved estimating the historical marginal tax rates on ordinar)' income and capital gains. Tbis paper uses the marginal tax rates provided in Exhibit 1 of Siegel and Montgomery [Winter 1995J. Because tax rates are heterogeneous, they chose an arbitrary single taxpayer earning $75,000 in "earned" (noninvestment) income in 1989 dollars. This level of income was deflated (inflated) by the Consumer Price Index (CPI) for earlier (later) years. They argue tbat this investor w ould be typical of individuals with sizable investment portfolios subject to tax. Because our data starts in 1976, we use the Siegel and Montgomery marginal tax rates on ordinary income and capital gains from 1976 through the end of their study in 1993. For the years 1994 through 2000, we use tax code information on the ordinar\' income and capital gains rates and adjtist earned income by tbe CPI for each year. After-tax returns for a given mutual fund in a given year are eomputed by adjusting the total return for the taxes that would have been paid on the dollar income and capital-gain distributions for tbat year.' Tbere is a slight upward bias in this aftertax return computation because xMorningstar includes both sbort-term and longterm capital gains in its yearly dollar-pershare capital-gain figure. The sbort-term

The mutual fund data used in this study came from tbe January 2001 Morningstar Principia Pro Plus for Mutual Funds, This database contains historical information on virtually the entire universe of mutual funds through the December 31, 2000, year-end, "fhe funds are classified by investment objective, and total return, income and capital gain distributions, annual expense ratios, fund size and turnover infonnation are provided for each of the funds along with a wealth of other fund data such as load status. There are over 12,000 funds included in tbis database, wbich is one of the most widely used research sources for mutual fund information. This study groups tbe funds into eight broad investment categories: Aggressive Growth ik Growth (AGG), Growth/ hicome Hi Fquity/Income (GIEl), Specialty Equity (SP), Small Company Equity (SCE), Corporate Bond (CB), Government Bond (GB), International Stock (IS) and Asset Allocation & Balanced (AAB). The comparison index funds include Vanguard Index 500, Vanguard Total Bond Index,

The methodology employed to test the hypothesis of significant differences in mutual fund returns involve a comparison of the cross-sectional mean (median) mutual fund returns each year and over the 25-year period. These tests are performed across investment classifications. The returns relationship is investigated by segmenting tbe sample by fund category and

Journal of Financial Planning/Septennber 2002

Contributions

Fortin

T A B L E

1 A

Summary Descriptive Statistics for Mutual Fund Categories

If Category | Fund Category 1
Years Index

Variable N Mean ^ Standard Deviation ^ledian

Total Return 12116 16.86 22.43 17.14 6644

ATTotal Return 12089 14.80 22.28 14.85 6634 11.71 13.56 12.46 3603 14.37 33.54 12.58 3968 14.65 26.03 12.83 7038 4.16 8.53 4.81 4245 4.56 6.11 4.98 4818 9.56 35.51 5.55
"inirnFuwniiiMiiiiiiYtihi ii n ni >

Turnover 9919 99.49 118.10 75.00 5566 68.05 55.63 56.00 3038 126.05 3S2.55 63.00 3246 90.23 93.53 72.00 5802 ' 130.07 137.97 93.00 3618 '' 170.31 197.36 117.00 3381 90.66 74.25 74.00 3192 92.07 94.88 80.00

Expense Ratio 10956 1.41 0.73 1.25 6045 1.28 0.99 1.11 3355 1.65 0.65 T.57 3591 1.52 0.86

Net Assets 13663 503.50 2504.73 62.90 7259 640.26 2691.11 71.70 4208 246.40 745.40 43.20 4492 206.16 557.11 48.90

AGG 1977-2000 Vanguard Index 500 GIEI 1977-2000 Vanguard Index 500 Spec. Equity 1977-2000 Vanguard Index 500
SCE

2 Years Index

pviean Standard Deviation Median

N

13.95 13.76 14.70 3606 16.03 33.97 14.24 3985 16.47 26.33 14.30 7047 6.48 8.60 6.85 4248 6.56 6.25 6.98 4825 10.79 35.58 6.81 4069

9
Years

Mean Standard Deviation Median

N

Index

1976-2000
Index Vanguard Small Cap Index Corp. Bond

Mean Standard Deviation ^ Median

5
Years Index

FN
Mean Standard Deviation Median
N

6393 1.04 0.50 0.92 3917 1.02 0.51 0.91 3633 1.89 0.67 1.85 3521 1.36 0.58 1.25

7820 275.07 817.73 57.80 4614 358.65 1102.15 56.80 5186 307.21 1580.08 31.55 4464 405.03 1648.97 50.30

1987-2000 Vanguard Total Bond Index Gov't Bond 1987-2000 Vanguard Total Bond Index Int'f Stock 1997-2000 Vanguard Total International Index AA Balanced 1993-2000 , Vanguard Balanced Index

6
Years Index

Mean Standard Deviation Median N Mean Standard Deviation ^Median

3 Years Index

4059 8.83 11.56 9.51

Years Index

Mean Standard Deviation Median

10.81 11.69 11.53

HtAe: N = number of nonmjssjiig fund/year obiervai

Journal of Financial Planning/September 2002

Contributions

Forti

TABLE

IB

Summary Descriptive Statistics Index Funds

Category
Index

Fund Category Vanguard Index 500

Variable N Mean Standard Deviation Median

Total Return 24 15.52 14.21 18.06 25

ATTotal Return 24 13.58 14.14 14.66

,

f< Expense

Turnover 23 13.09 10.26 8.00

J Ratio 23 0.26 0.08 0.22 24 0.82 0.58 0.94

Net Assets 24 16,731.37 30,659.00 1,429.45 25 677.02 1,156.12 37.10 4 2,186.50 821.76 2,507.50 7 0.20 0.00 0.20 13 0.20 0.04 0.20 8 1,519.68 1,259.03 1,038.30 14 3,146.17 3,656.42 1,435.45

Index

Vanguard Small

24 11.41 18.49 15.23 4 6.70 19.70 6.79 J 8 11.57 10.59 12.50 14 5.52 5.87 6.23

24 ^ 70.33 39.09 75.50 3 3.00 2.65 2.00

7 23.71 7.73 25.00 13 42.23 14.94 39.00

Cap Index

rMean Standard Deviation Median

14.18 18.76 18.1S 4 7.28 19.75 7.41 8 12.84 10.70 13.78 14 7.93 5.93 8.62

Index

Vanguard Total International Index

N Mean Standard Deviation Median

Index

Vanguard Balanced Index

N Mean Standard Deviation Median

Index

Vanguard Total Bond Index

N Mean Standard Deviation Median

Note: N = number cif nonmissing fund/year ob^ervat

return, turnover, expense ratio and size variables for the funds across investment classifications. The second column of Table 1A provides the comparison index fund for eaeh category and the years used for the analysis. Notice that each of the category coiTiparison periods is different due to differing lives of the respective index funds, so the following discussion should be read with that in mind. For example, the Vanguard Index 500 has 24 etJTTiparison years because the fund \v as started in 1976 and Momingstar reports its fii-st full year return in 1977, while the Vanguard Total International Index fund has been in existence for on!v four full \-ears.
Journal of Financial Planning/September 2002

Continuing with Table lA, the total return/after-tax total return was highest for the equity funds and lowest for the bond funds. The Aggressive Growth & (Growth (16.86% / 14.80%) and Small Company (|uity (16.47% / 14.65%) funds were the leading equity funds. The lowest total return category was Coqiorate Bond funds (6.48% / 4.16%). For the comparison index funds, the Vanguard index 500 fund (Lv52% / 1 3.58%) had the highest total return. The Vanguard Total International Index fund (7.28% / 6.70%) and the Vanguard Total Bond Tndex fund (7.93% / 5.52%) had the lowest total returns.

Growth/Ineome & Kquity/Income funds had the lowest turnover (68.05 percent), while Government Bond funds {170 percent) and Specialt)' Equity funds {126.05 percent) had the highest turnover. As anticipated, the index funds had lower turnover ratios than their comparison fund categories. The Vanguard Small Cap Index fund (70.3 3 percent) had the highest turnover, while the Vanguard Total Internationa! Index fund (3.00 pereent) had the lowest turnovur. Uncxpeetedly, the Vanguard Total Bond Index fund (42.23 pereent) had a relatively higher turnover.

Fortin

Conlributions

capital-gain distributions should he subject to the higher ordinary income-tax rates, but it was not possible to make this adjustment. Consequentiy, the differences between before- and after-tax returns presented in this article are slightly smaller than would actually be expected.

for this year but did not do so for the Vanguard Index ,SO(I until 1977, The number of observations is the same for both indexes for after-tax total returns because

the lagged prior year net asset value is required to make this computation. Tables lA and IB provide a broad overview of the total return, after-tax total

Table lA provides summary descriptive statistics for each of the eight investment categories and Table IB provides similar information for the five comparison index funds. The expense ratio is the total fund expenses expressed as a percentage of fund assets. Total fund expenses include all management and administrative fees, 12b1 distribution fees, and all other assetbased costs incurred by the fund, except brokerage costs. Sales charges (front or deferred loads) are not included in the expense ratio. Fund size is the total net assets of the fund (in millions of dollars). Turnover, as defined by the Securities and Exchange Commission (SEC), is computed by taking the lesser of purchases or sales (excluding all securities with maturities of less than one year) as a percentage of average fund assets. Return is the yearly total return computed based upon changes in fund net asset value (NAV) assuming reinvestment of all income and capital gain distributions. Returns are net of the funds expense ratio and brokerage costs, but are not adjusted for front or deferred loads. The number of observations (N) is the total number of non-missing observations for each variable. N is often different for the five variables because data is missing in the iMorningstar database more often for some variables than others. In Table IB, the number of observations for the Vanguard Index ,>00 total return variable is 24, while it is 25 for the Vanguard Small Cap Index. This is because our analysis started in 1976 and Morningstar reported a total return for the Vanguard Small Cap Index

Financial Planning Professional

If you offers:

need

Financial Planning software that

(check all that apply) ^ comprehensive OR modular planning Ef integrated input and analysis forms [^ flexiblity ^ customizable and editable reports Ei unsurpassed technical support and customer service Ef ease of data entry for your or your assistant Then you'll Lumen Systems' Financial Planning Professional.^ "
IS

want

all that nd more!

visit

umen Systems at
www.lumensystems.com or call 800 233-3461

Journal of Financial Planning/September 2002

Contributions

Paired Comparison T-StatisticsFund C

Comparison Using Total Return Variable
Category 1 Years AGG

mpared to (Index-Active Fund)

Comparison Using After-Tax Total Return Variable

s iJfiiM'v^n.;:^^'^^^"'^"'^'^ 1 _ Category Mean Std.Error Prob. 12,116 6,818 GIEl 1977-2000 6,644 4,384 Spec Equity 1977-2000 3,606 2,191 Median Sign Test* Mean Std. Error Prob. Median Sign Test* Mean Std. Error Prob. Median Sign Test* '""""" 0.0001 0.031 0.573 0.9575 5.78 0.0001 Years N No. Pos. 1977-2000 3,603 2,213 No. Pos. 9 ^~_ _ -0.928 0.168 0.0001 1.59 0.0001 2.828 0.143 Years N 1 AGG

IHHIH^^'Sua'^rdrrrdtxiSo"'
Mean ~-..--. 0.121 0.167 ^ Std. Error 1977^000" Prob. 12,089 7,178 GIEl 1977-2000 6,634 4,649 Spec Equity Median Sign Test* Mean Std. Error Prob. Median Sign Test* Mean Std. Error Prob. Median Sign Test* Variable Mean Std. Error 0.0001 0.752 0.567 " 0J847" 6.47

lV47bi~
2.66 0.0001 3.983 0.144

N
No. Pos.

. NOJ^S. i
2 Years

2
Vears" N No. Pos.

"""b.oooi

O.oobT^

9
Years N No. Pos. Category 8

o.oobf"
Vanguard Small Cap Index

Years N No. Pos. 'Category

1976-2000 3,968 1,806

Prob. Median Sign Test* Variable

0.96 0.093
0.0001 0.78 0.0001 Government Bond

Corp. Bond

Mean Std. Error Prob. Median Sign Test Mean Std. Error Prob. Median Sign Test

1.135 0.093 0^001

Years N No. Pos. 6 Years N No. Pos.

1987-2000 7,038 4,327 Gov't Bond 1987-2000 4,24S 2,826

6.000?
0.841 6.050

d.ooo

Variable International Stock Years N 1997-2000 4,825 2,328 Mean Std. Error Prob. Median Sign Test* Variable AA Balanced Mean Std. Error Years
N N0.P0S.

Category 3

Int'l Stock

No. Pos. 1
Category 4

1997-2000 4,818 2,395

Category 4 Years N No. Pos, Variable AA Balanced 1993-2000 4,059 2,746 Mean Std. Error Pi-ob. Median Sign Test 2.47 0.0001

1993-2000 4,069 2,572

Prob. Median Sign Test*

Sign test one-taiied probabilitythatthediffeienceissignificifntly positive (Index-Active Fund), NS

Journal of Financial Planning/September 2002

ConlribuLions

for

tin

T A B L E

3 A

Paired Comparison T-TestFund Category Compared to Mean {Index-Active Fund): Comparison Variable:Total Return
AA Balanced

Notes Indicates significance a! 5% level or bettpr

Expense ratios were highest for Intci'iiational Stock funds (1.89 percent) and Spccialt)- Equity funds (1.65 percent) and lowest for Government Bond funds (1.02 percent). As expected, the comparison index funds had, o\'cral!, low er expense ritios. The Vanguard Small Cap Index fund (0.82 percent) had tlie highest expense ratio and the \'an"uard 'i"otal Bond Index
Journal of Financial Planning/September 2002

fund (0.20 percent) and the X'anguard Balanced Index fund (0.20 percent) had the lowest expense nitios. The largest funds were Growth/Income ik FAHiit\7income ($640.26 million) and Aggressive CJrovvth & (iroM-th ($.S()3.5() million). The smallest category was Small Company LAjuity funds ($206.16 million). The index funds w ere generally much

larger than their peer comparison funds. The largest index fund was tlic \'anguard Indcv 500 ($16,7:^1..S7 million) and the smallest was the Vanguard Small Cap Index fund ($677.02 million). Table 2 presents the paired comparison t-tests for total return (left panel) and aftertax total return (right panel) for each of the eight fund categories overall, with respect

Contribntions

Fort in

T A B L E

3 B

Paired Comparison T-TestFund Category Compared to Mean (Index-Active Fund): Comparison Variable: After-Tax Total Return
Fund Category Category No. 1977 1978 1979 1980

AGG
1
-13.250* -10.090-16.220* -8.020* -3.660* -7.510^ -0.680 9.317' 0.118 3.163^ 5.825* 0.697 4.154^ 2.070^ -9.737 * -1.059' -2.048' 3.373' 6.847' 5.537* 10.691* 8.615*

' GIEI

Spec. Equity

SCE
8 -8.900* -19.460' -1.188 2.015 2.032 19.289^ -10.576^ -18.348^ -7.703^ -10.951 ^ -4.015* 4.821 ' -15.485^ -7.790* -3.020 4.279 1.861^ -0.236 -1.408 -2.502* 2.840* -3.830^

Corporate Bond 5

Gov't Bond 6

Int'l Stock 3 na na

AA Balanced
4

2
-1.242 -3.979* 5.577' -5.257* -2.582* -0.379 2.190* 3.430*
2.373

9
-20.088 *

na na
na

1;

na
na na
na na

na na na na na na

-5.463 ' -52.949^ -9.725 3.786 -3.026 2.812 9.792 ^ 9.688* -3.165 -0.587 6.995* 3.5734.539' 2.730 0.555 -16,705* 4.57211.002' 4.591 ' 18.201 ^ 19.949* -11.361* -14.498*

na na na na na
na

na na na na na na
na 0.615 -2.526* 5.954* f 8.037* -7.538* -2.928* -2.206* 0.703* 1.728* -2.577* -0.351 * 3.620* -2.940* 1 7.795*

f 1981
1982

na na
na na na 0.309 -0.086 1.034* 0.163

1 li
c

198S 1984 1985 1986 1987

na
na

na
na na

na
na na na

6.510* 0.607 : 8.623* ; 2.711* 2.316* /. -0.698 -1.321* 3.171* 7.603* 4.377*

sari:

1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000

na

na
na na
na
n a '

to their comparison index funds. We take the index fund return minus the active fund return and then compute a paired comparison t-test based upon this value. Therefore, a positive mean return indicates that the index fund outperformed the average actively managed fund. The statistical results performed on the differences in returns appear to reflect skewed distribuJournal of Financial Planning/September 2002

liuo

1 1 1

0.924'
0.888'

na
na
-1.712* 1.329* 4.772* 1.809* 5.920* 5.783* 3.828* -2.474*

1.558*
0.426 *

na na na
na -2.524* 11.054" -21.843* 2.962'

2.649*
0.423'
1.089*

i
1

7.939*
14.921 "

1.143' -0.271 0.820*

-n.on*
-3.003*

11.311*
-9.619*

1 -15.506*
-5.565'

Notes; 'Indicates significance st S% level oi bellsi

tions when comparing the mean and median returns values. We, therefore, compute the nonparametric sign test to determine if there is a significant difference in the number positive versus negative differences in returns (Index minus Active fund). All fund categories, except Specialty Equity funds, showed a significant difference in returns (at the .0001 level) between

the index funds and the actively managed funds for total return. Interestingly, three equity eategories exhibit a negative relationship for total return (and two categories for after-tax total return), indicating that the actively managed funds significantly outperformed the index funds over the full comparison period. The Aggressive Growth & Growth (total return only).

Fortin

Contributions

Small Company H!quity and International Stock funds exhibited this nesjativc relationship (note tiiat the IS funds only contained four years of data). After-tax total returns exhibited similar relationships, except that the AGG funds were not significant (and positive), instead of significantly negative. The nonparametric sign test provides a stronger indication of these results. All categories, except two {Small Company Fxjuity and International Stock funds), .show significantly more positive than negative returns (based on a one-tailed test at the D.OOO 1 level) for both total return and after-tax total return. 1 he nonparametric results provide additional evidence that index funds outperform activel)' managed funds for all categories, except for SCE and IS funds. Tables 3A and 3B present the results of the same paired comi)arison t-test for each ot the eight fund categories on a year-byyear basis for total return and after-tax total return. Note that the SCE comparison starts in 1977 rather than 1976 to simplify the table. Over all categories and time periods, 1 i 3 out of 136 differences in total returns and 110 out of 136 after-tax total returns are significant. Table 4 shows how many total returns arc significantly positive (index fund outperforms) versus significantly negative (actively managed fund outperforms). Five ofthe eight total return categories show that index funds outperformed actively managed funds in the majority of years (the IS category was split), while active management outperformed indexing in the majority of years in two CiUei;;t)rics (SCE and CB). These results \\-cre similar for after-tax returns except for the ACXi category. This suininar)- reinforces the conclusion that Small Company Equity funds outperform rhe index funds whether considering total returns or after-tax total returns. Fortin ;ind Miehelson (1999) also found similar total return results for Small Com-

Tota) Return AGG Positive Negative 12 11 GIEl 13 8

SP
SCE

CB

GB

IS

AAB

10 6

5 13

10 0

^Tax Total Return

I I AGG
Positive Negative 10 11

GIEl
14 6

SP

SCE

10 6

pany F-quity funds. The results for the International Stock funds were split (two positive and tv\(> negative) on both a before- and after-tax basis, although overall in Table 2, active management outperformed on both .1 pre-tax and after-tax basis. The International Stock fund results should be viewed with caution given the relatively short time periods involved. In addition, it appears that the results in Table 3 are sensitive to the biisine.ss cycle. The managed funds seem to outperform the index funds during periods hen the economy is either going into or out of a recession. There are negative values of differenees in returns (index return minus active fund return) fairly consistently during 1979-82, 1991-93 and 1999-2000. It appears that active fund manatjcment is better than investing in itidex funds when guiding portfolios through difficult times. This is an important point to consider w hen making investment decisions because the overall results presented in I able 2 hide this tendency. We understand tbat there is an obvious sur\'ivorship bias in these results. Because our study period encompasses 2.S years, there will be a number of mutual funds that perform poorly and disappear from our sample in tlie later years. As such, this bias favors acti\e fund management over indexing and any indexing (active) advantage would be greater (less) than reported.

This paper's primary contribution is in pro\iding more conclusive evidence on the debate on the benefits oi mutual fund uidexing and whether actively managed funds perform as well as index funds. An important feature of this research is that we analyze the results for both total return and after-tax total retum. We find that, on average, index funds outperform actively managed funds for most equity and all bond fund categories on both a before-tax and after-tax basis. However, actively managed Small Company Equity (SCE) funds and International Stock (IS) funds significantly outperform the index over most ofthe study period. .Managers of these funds appear to be able to invest to take advantage of mispricuig in these presumably less efficient markets. The nonparametric tests further reinforce these results. The Sign Test establishes that index funds outperform activel)- managed funds for al! categories, except SCE and IS funds. The overall results should be view ed with caution, howex'cr, as there is evidence that actively managed funds outperform the index funds during periods when the economy is either going into or out of a reeession. It appears that active fund nianaticnient is better than index funds at guiding portfolios through rough times. A number of caveats are also In order. First, tbis studv tines not consider either
Journal of Financial Planning/September 2002

LoiiliMbulions

Fortin

front or deferred loads, and the relative performance of index mutual funds versus actively managed load or no-load funds is an empirical question left unanswered by our study. Second, the methodology in this study does not take into account a risk/return trade-off. We onlv examined returns b)' categor\", and a number of funds with lower returns might actuall}' have a better risk/return trade-off than the market because they have lower risk than the market. Third, our tests do not take inio account persistence of fund superiorit}- oxer the passive index fund. An}' fund that consistently outperforms the market index tends to be averaged out in our cross-sectional methodology. There is a large body of literature in this area [see Zheng (1999)1 ^md this study docs not explicitly address this issue.

3. More specifically, the ending net-asset value from the prior )-ear is multiplied by one plus the decimal retum for the current year. 'Fhen the computed tax on the ordinary income and capitalgain distributions for the year are subtracted from this appreciated net-asset value to give an after-tax net-asset value. After-tax total returns are then computed by dividing the after-tax netasset value bv the ending net-asset value from the prior year and subtracting one.

nn

Endnotes
1. N(ite that the total number of mutual funds represents totals (ner different time periods. For example, AGG funds were examined over 1977 through 2000 and IS were reviewed over the years 1997 through 2000. 2. The index funds represent specific segments of the market. The Vanguard Index .SOO fund reflects the S&P 500 Index, the Vanguard Total Bond fund reflects the Lehman Brothers Aggregate Bond Index, the Vanguard Small Cap fiind reflects the Russell 2000 Index, the Vanguard 1 otal international fund refiects the MSCl European, Pacific and Emerging Markets Indices and the V'anguard Balanced fund reflects the combination of the Wilshire 5000 and the Lehman Brothers Aggregate Bond Index.
Journal of Financial Planning/September 2002

Arnott. R., A. Berkin and j . Ye. "How Well Have Investors Been Served in the 1980s and 1990s?" The Journal nf Pnrtfolio Aianagemeiit. 26, 4 (Summer 2000): 84-91. Bogle, 1. "The 'Friumph of Indexing." 'Fhe Vanguard Group (April 1995): 1^5. Bogle, J. "Be Not the First...Nor Yet the Last." The Vanguard Group, from a speech presented May H, 1996, at the 1996 AIMR Annual Conference in Atlanta, Georgia. Bogle, J. "What Can Active ^Managers Learn from Index Funds?" 'Fhe Vanguard Group, from a speech present to the Bullseyc 2000 Conference in Foronto, Canada, December. Elton, E.J., M.J. Gruber and C.R. Blake. "'Fhe Persistence of Risk-Adjusted Mutual Fund Performance."_/ourn.i/o/^' Business. 69, 2 (1996: 1.^3-157. Fortin, R. and S. Micheison. "Mutual Fund Indexing Versus Active Management." journal of Financial Phnning. 12, 2 (1999): 74-81. Gruber, M.J. "Another Puzzle: Fhe Growth in Actively Managed Mutual F'unds." T h e JiHirnal of Finance. 5 1 , 3 (July 1996): 783-810. lx)wenstein, R. "Indexing's Real Uncertainty Principle." The Wall Street Journa/(January 30, 1997): Cl.

Malkiel, B. "Not So Random." Barron's (April 22, 1996): 55. Malkiel, B. "Returns from investing in Equity Mutual Funds 1971 to 199L" The Journal of Finance. 50, 2 (June 1995): 549-572. Morningstar Mutual F'unds. Mutual Funds OnDisk, operations manual, Chicago, Illinois: January 1996. Siegel, B. and D. Montgomery. "Stocks Bonds and Bills after Taxes and Inflation."^our/i;?/ of Portfolio Management (Winter 1995): 17-25. Wermers, R. "Mutual F'und Performance: An Empirical Decomposition into Stock Picking Talent, Style, Transactions Cost and Expenses." The Jounial of Finance. 55, 4 (August 2000): 1,655-1,703. Zheng, L. "Is Money Smart? A Study of Mutua! FVnd Investor's Fund Selection Ability." The Journal of Finance. 54, 3 (]une 1999): 901-933.