Smoothing the Path (The Sum is Greater Than The Parts)

Reprinted from Financial Planning Magazine, October 2007
Craig L. Israelsen, Ph.D.

he vast majority of studies that compare the performance of actively managed funds and index funds tend to focus on individual funds. I would suggest that such studies should also consider differences between active and passive management at the portfolio level. Let's face it, the natural habitat of any fund, whether passive or active, is within a portfolio of several funds. Isn't this precisely what financial planners want clients to focus on-the performance of the portfolio as a whole, rather than each part?

Having said that, comparing portfolios is a complex task, made more so by the fact that actively managed funds and index funds are philosophically different products. By definition, index funds attempt to mimic (to a greater or lesser degree) a given index. As a result, they are pure asset classes with either an all-stock or all-bond portfolio, and cannot deviate to take advantage of (or defend themselves from) changing markets. Actively managed funds, on the other hand, have greater latitude in the number and types of assets they invest in. For instance, many U.S. stock funds may also have a small amount of cash, a small dose of bonds and/or preferred stock, and some exposure to international equities.

Considered in that light, an index fund is generally a poor choice as a stand-alone investment. Said differently, an index fund is a poor choice for a person who is investing in only one fund because most index funds lack broad diversification.

On the other hand, there are many actively managed asset-allocation and balanced funds that would be good choices as a stand-alone fund because of their exposure to a broad range of asset classes (US large and small stock, non-US stock, bonds, cash, and other specialty assets). The market seems to agree: As of December 31, 2006 there were 297 distinct funds in Morningstar's "Moderate Allocation" category. Of that total, 296 were actively managed funds and only one was an index fund (Vanguard Balanced Index).

At that time, the average portfolio composition of Moderate Allocation funds was 51.3% U.S. equities, 8.8% non-U.S. equities, 27.9% bonds, 8.7% cash and 3.4% other. So, the casual advice "just put your money in an index fund and let it ride…" is poorly conceived IF the investor chooses only one index fund (Vanguard Balanced Index being the exception). Index funds have tightly defined portfolio holdings-which means that they need a supporting cast of other funds (index or otherwise) to produce a portfolio with sufficient diversity.

This study examines the performance of a portfolio of four popular index funds, and compares it to the performance of a portfolio of four actively managed funds (using aggregated performance data of popular actively managed funds). This approach is much like studying the behavior of family members within their family setting, rather than studying each person in isolation.

The data for this study were obtained from Morningstar Principia. The four index funds chosen to represent the passively managed, index-based portfolio were Vanguard's 500 Index, Extended Market Index, Total International Stock Index and Total Bond Index. Funds, rather than raw indexes, were chosen because raw indexes are not purchasable. Moreover, raw indexes do not have expense ratios, which is an unrealistic assumption in the real world of actual products purchased by actual investors. Vanguard is clearly the leading provider of index funds, offering 114 products (index funds, enhanced index funds and exchange traded funds) that collectively possessed nearly 39% of the $1.4 trillion invested in index funds, enhanced index funds and ETFs as of December 31, 2006.

The Vanguard Total International Stock Index (VGTSX) is comprised of three Vanguard funds (Pacific Stock Index, European Stock Index, and Emerging Markets Stock Index). The bulk of the fund tracks the MSCI EAFE (Europe, Australasia, and Fare East) index, while a portion of it attempts to track the MSCI Emerging Markets Index. Finally, Vanguard's Total Bond Index (VBMFX) tracks the Lehman Brothers Aggregate Bond Index.

The performance of each index fund over the ten-year period from 1997 to 2006 is shown in "Index Portfolio". In addition to the annual performance figures, the annualized returns over 3, 5 and 10-year periods have been calculated as well as standard deviation of return over the same time frames.

Index Portfolios

In the right-hand portion of "Index Portfolio," you'll find the performance of the four index funds as a portfolio. Three different portfolios were simulated: Aggressive, Moderate and Conservative. The aggressive portfolio was comprised of 40% Extended Market Index fund and 20% in the remaining three index funds. In the moderate portfolio each of the four index funds had an equal weighting of 25%. The conservative portfolio was weighted 50% in VBMFX, with 20% committed to VFINX, 15% in the VEXMX and 15% in the VGTSX. The portfolio was rebalanced at the start of each year. The results shown are pre-tax and assume no commissions were paid.

Active Portfolios

Selecting actively managed equity funds that were comparable to each of the four index funds was more challenging. The approach used here was to select the largest funds (in terms of net assets) in each of the four categories (large U.S. equity, mid/small U.S. equity, non-U.S. equity, bonds) at the end of 1996. The performance data for the active portfolios are reported in "Active Portfolio".

The 10 largest large cap blend funds at the end of 1996 were (using their names in 1996) Fidelity Magellan, Investment Company of America, Fidelity Growth & Income, Fidelity Contrafund, Janus, Dean Witter Dividend Growth B, Growth Fund of America, Fidelity Blue Chip Growth, T. Rowe Price Equity Income and Fundamental Investors. These 10 funds all survived the subsequent ten-year period from 1997-2006. The combined net assets in these ten funds at the end of 1996 was $198 billion, or 46% of the total assets of all the U.S. equity funds classified as residing in the large cap blend style box at the end of 1996. Thus, this small sample of large cap blend funds represents a very large share of total large-blend assets in 1996.

It is true that some of these 10 large blend funds style drifted into large value or large growth in the ensuing decade, but we only now know that in hindsight. This analysis seeks to replicate what an investor's actual choices were in 1996-that is, to invest in either an index fund (as represented by VFINX) or an actively managed fund in the same style box (as represented by the 10 largest funds at the end of 1996). There is no survivor bias among these ten funds because they were selected according to asset levels at the start of this 10-year analysis period, not the end.

The group of actively managed funds representing mid and small cap U.S. equities was comprised of the largest three funds in each of the six appropriate style boxes (mid value, mid blend, mid growth, small value, small blend, small growth) at the end of 1996. Therefore, the aggregate performance of 18 funds was used to compare against the performance of VEXMX. The largest 18 funds in these six categories included (using their names in 1996) Fidelity Value, Pioneer II, Mutual Beacon, Fidelity Retirement Growth, Acorn, Mutual Discovery, AIM Constellation, Putman Voyager, Brandywine, Fidelity Low-Priced Stock, Heartland Value, T. Rowe Price Small Cap Value, Lindner Growth, Vanguard Explorer, Baron Asset, Dreyfus New Leaders, Kaufmann, AIM Aggressive Growth and Keystone Small Company Growth S-4. (Nineteen funds are listed here because Lindner Growth did not survive the entire period.)

Although AIM Aggressive Growth merged into AIM Constellation in March of 2006, we nevertheless utilized the actual year-to-year returns of AIM Aggressive Growth from 1997 to 2005 in this analysis. Keystone Small Company Growth became Evergreen Mid Cap Growth B, but its performance history remained intact. Other name changes included Fidelity Retirement Growth's becoming Fidelity Independence, and Pioneer II's becoming Pioneer Value.

One of the largest funds in 1996 that did not survive the ten-year period was Lindner Growth Investors. Therefore, it was replaced with the next largest fund in 1996 in the small cap value space, namely T. Rowe Price Small Cap Value. However, the annual returns of the Linder Growth Investor up through 2003 were included in the analysis so as to eliminate any survivor bias.

As of December 1996, these 18 funds (19 counting Lindner Growth) accounted for about $80 billion of the $322 billion of the net assets in the 1,115 funds in their six style boxes-or 24% of the total mid and small cap assets. As with the 10 largest large cap blend funds, this small sample of 18 mid and small cap funds represented a large percentage of the mid and small cap assets at the end of 1996.

The methodology of this study equally weights the returns of each of the 18 funds within each of the six style boxes (mid value through small growth). This is significantly different from the VEXMX Index fund which represents the return of a market-cap weighted index (either the DJ Wilshire 4500 or the S&P Completion Index). Market-cap weighting will obviously increase the impact of the larger mid cap stocks and decrease the impact of the smaller small cap stocks.

The 10 largest actively managed international funds in 1996 were selected from three Morningstar categories: Foreign Stock, Diversified Emerging Markets and Diversified Pacific Stock. The 10 largest funds were American Funds Euro Pacific A, Bernstein Tax-Managed International, DWS International (i.e., Scudder Intl), Fidelity Overseas, GMO Intl Intrinsic Value III, Harbor International, T. Rowe Price Intl Stock, Templeton Developing Markets, Templeton Foreign A and Vanguard Intl Growth. All 10 funds survived the entire 10-year period. At the end of 1996, these funds held $61 billion of the $143 billion among all 516 funds in the three categories-or 43% of the assets.

Finally, the 10 largest actively managed intermediate-term bond funds in 1996 were PIMCO Total Return, Bond Fund of America, General Electric S&S Long-Term, IDS Bond A, Standish Fixed Income, T. Rowe Price New Income, Bernstein Intermediate Duration, Fidelity Investment Grade Bond and Intermediate Bond Fund of America. These 10 bonds managed 40% of the $88 billion held by all 390 intermediate-term bond funds at the end of 1996. All 10 funds survived the entire 10-year period to 2006.

The Verdict

Each of the three index portfolios (aggressive, moderate, conservative) outperformed the actively managed portfolios over the recent 3-year and 5-year periods, but underperformed the actively managed portfolios over the 10-year period. Interestingly, in every case, the return volatility (i.e., standard deviation of returns) of the actively managed portfolios was lower than that of the index portfolios. This was also true at the individual fund level.

The year-to-year performance differential between the aggressive index portfolio and the aggressive active portfolio is shown in "Downside Protection". The 10-year annualized returns are similar for both: 8.77% for the index portfolio vs. 9.19% for the active portfolio. However, the year-to-year performance behavior reveals a pattern in which the index-fund portfolio outperformed when the U.S. equity market was buoyant (1997 to 1999 and 2003 to 2006), and underperformed when the most of the US equity market was in a funk (2000 to 2002). Similar year-to-year performance differences between the active portfolio and the index portfolio were observed in the moderate and conservative portfolios, but at lower levels of overall return.

By design, equity index funds are fully committed to equities. As a result, equity index funds (i.e., index-fund based portfolios) will likely outperform actively managed funds when the equity market is experiencing a broad-based rally (see "Cash Flow"). Why? They hold almost no cash, and therefore endure very drag on returns. Actively managed funds, on the other hand, may have sizeable cash drag.

When the equity market is struggling, however, actively managed equity funds enjoy latitude (in varying degrees) to seek shelter in diverse asset classes. Whether or not such latitude adds value-above and beyond the performance of a passively managed index fund-is clearly at the heart of the active/passive debate.

The results of this study clearly indicate that latitude at the helm (i.e., potentially broader asset diversification among actively managed portfolios) offers an opportunity for protection during bear markets. And because losses tend to rattle investors, the potential downside protection in a portfolio of actively managed funds during a bear market may more than offset its likely underperformance during bull markets.

Finally, as shown in "Fluctuating Active Premium", there is considerable fluctuation in the advantage associated with active management. On average, actively managed U.S. large cap funds (noted by the pink line) outperformed the Vanguard 500 Index fund by about 25 basis points over this 10-year period.

Actively managed mid and small cap funds (blue line) underperformed the Vanguard Extended Market Index Fund in seven of the ten years. The average discount (or active management underperformance) was 57 basis points, and the median underperformance was 220 basis points. When actively managed mid/small cap funds outperformed, the premium was sizeable (2,016 bps in 2000, 536 bps in 200, and 77 bps in 2002). Recall that the aggregate performance of these 19 mid/small cap funds does not reflect the same degree of market cap weighting as exists in the VEXMX fund. In other words, the performance of the small cap and mid cap funds were equally weighted in the active portfolio-not the case in the VEXMX fund.

Actively managed international equity funds (yellow line) outperformed the Vanguard Total International Stock Index fund in six of the 10 years. The average active management premium was 146 basis points, while the median premium was 244 basis points. As was the case among actively managed U.S. large cap and mid/small cap funds, actively managed international funds outperformed index-based funds during the 2000 to 2002 slump.

Finally, actively managed bond funds (green line) underperformed the Vanguard Total Bond Index by an average of 13 basis points each year (median differential was 0 bps). The differential in performance between index-based and actively managed bond funds is far less dramatic than among equity funds.

Active management can be advantageous when the sea is rough (such as during the stormy years of 2000-2002), but when the equity waters are calm its hard to beat indexing. In other words, the potentially eclectic mix of assets in some actively managed funds can be a real plus in certain years. On the other hand, the precise portfolio focus of index funds gives them an advantage when their slice of the market has a solid run (i.e., large cap US equity from 1995-1999). Active management and index-based passive management both have merits that manifest themselves during different seasons.

Most assuredly, the active vs. passive debate will continue into the future. However, the quality of the debate may improve if we focus on the performance of portfolios, rather than simply the performance of individual ingredients. As many have wisely observed, the whole is greater than the sum of the parts.

Index Portfolio

Index Funds	Vanguard 500 Index	Vanguard Extended Market Index	Vanguard Total International Stock	Vanguard Total Bond Index	Aggressive Portfolio	Moderate Portfolio	Conservative Portfolio
Best Fit Index	S&P 500	Dow Jones Wilshire 4500 (from 1997-2005)	MSCI World xUS	Lehman Brothers Aggregate	40% Ext Mkt  20% 500 20% Intl 20% Bond	25% in each index	50% Bond  20% 500     15% Ext Mkt 15% Intl
Morningstar Category	Large Blend	Mid/Small Cap Blend	Foreign Large Blend	Intermediate-Term Bond	Annually Rebalanced, Pre-Tax
Annual Performance					Annual Performance
1997	33.19	26.69	-0.78	9.44	19.05	17.14	15.24
1998	28.62	8.35	15.60	8.58	13.90	15.29	13.61
1999	21.07	36.22	29.92	-0.76	24.53	21.61	13.76
2000	-9.06	-15.51	-15.61	11.39	-8.86	-7.20	-0.79
2001	-12.02	-9.17	-20.15	8.43	-8.42	-8.23	-2.59
2002	-22.15	-18.06	-15.08	8.26	-13.02	-11.76	-5.27
2003	28.50	43.43	40.34	3.97	31.93	29.06	20.25
2004	10.74	18.71	20.84	4.24	14.65	13.63	10.20
2005	4.77	10.29	15.57	2.40	8.66	8.26	6.03
2006	15.64	14.27	26.64	4.27	15.02	15.21	11.40
Growth of $10,000	22,272	25,231	21,165	17,843	23,177	22,575	21,344
Annualized Total Return					Annualized Total Return
3 Yr Return (2004-2006)	10.29	14.37	20.93	3.63	12.74	12.33	9.19
5 Yr Return (2002-2006)	6.06	11.95	16.08	4.61	10.46	10.05	8.20
10 Yr Return (1997-2006)	8.34	9.70	7.79	5.96	8.77	8.48	7.88
Annualized Standard Deviation of Return					Annualized Standard Deviation of Return
3 Yr Std Dev (2004-2006)	5.44	4.21	5.54	1.07	3.57	3.64	2.82
5 Yr Std Dev (2002-2006)	18.74	21.96	20.50	2.17	16.19	14.80	9.29
10 Yr Std Dev (1997-2006)	19.12	20.98	21.30	3.76	15.14	13.81	8.51

Active Portfolio

Actively Managed Funds	US Equity Large Cap	US Equity Mid and Small Cap	International Equity	Bonds	Aggressive Portfolio	Moderate Portfolio	Conservative Portfolio
Fund Group	Largest 10 Large Blend funds in 1996	Largest 18 Mid/Small funds in 1996	Largest 10 International Funds in 1996