Seeking Alpha
Profile| Send Message|
( followers)  

With the Immortal 12-EFT Strategy my objective is to present a strategy to simplify a lifetime of investing while maintaining a good shot at outperforming the vast majority of popular funds. The strategy eliminates the need for constantly watching the market and worrying about it. It takes only a few hours a year! I will be tracking the portfolio and will post periodic updates.

I have presented another strategy for those who prefer a more hands-on approach with individual stocks.

There is no shortage of ETF portfolio strategies in financial literature online or in print. The primary goal of practically all of these strategies is to beat their corresponding benchmark index. To accomplish this, the vast majority use some form of market timing. However, study after study has shown that an overwhelming majority of actively managed funds (and by extension portfolios of funds,) underperform the indices over time.

The main reason most money managers underperform the market averages is ironically because they try to beat it on a short-term basis, usually quarterly or annually. Unfortunately they are always under pressure to be short-term results centric because that's what sells. So they tend to engage in short term tactics, primarily market timing, to boost performance, which often backfires. But individual investors are not beholden to anyone, so they can make investment decisions with long-term results in mind and not swayed by short-term ups and downs.

In this strategy I refashion and combine three common principles: buy-and-hold, buy low, sell high, and be in the market at all times. In all three cases I've attempted to avoid the pitfalls. In buy low, sell high. I don’t try to predict the market tops and bottoms or time the trades based on stock market levels; and in buy-and-hold I buy and methodically rebalance. So, despite some recent proclamations to the contrary, buy-and-hold is alive and well, at least in the case of funds.

The strategy would involve at most six trades a year (three round-trip). It is straight forward enough that a novice individual investor can adopt and execute it with little risk of crashing it, since it's run almost entirely on auto-pilot.

Adopting the strategy entails two main phases: portfolio formation and strategy execution.

Portfolio Formation

The benchmark for the portfolio will be the S&P 500 (NYSEARCA:SPY) Index. Forming the portfolio required deciding on the constituents of the portfolio, the optimal number of the constituents, and the allocation of each. Starting out with the right portfolio is essential to the strategy, so I spent a lot of time on it. In picking the ETFs I started top-down. First, I asked what sectors are the best to be in, and secondly, which ETFs are the best choices within each sector.

For sector selection I used two principles in particular as my guidelines — diversification and long-term investment themes. For diversification I considered large cap / small cap, domestic / foreign, correlation factors, and of course sector variation. Some correlations that may appear excessive, such as small caps and emerging markets, are intentional, emphasizing the importance of those sectors.

For long-term themes I considered themes that are bound to influence markets in medium and long term. Some of the themes that went through my thought process include:

  • Demographic trends
  • Energy consumption
  • The shift of power (west to east)
  • Healthcare
  • Technology
  • Food production / Agriculture
  • Industrial commodities

For choosing the best ETF within each sector I considered several factors including, the past performance, the index being tracked, the number of holdings and concentration, the size, the volume, the turnover rate, and the expense ratio.

From the outset I decided to exclude closed-end funds (CEFs) due to their active management, and exchange-traded notes (ETNs), since in general they are susceptible to the credit risk of their issuers.

Why did I settle on 12 ETFs? Keeping diversification and the aforementioned themes in mind, I wanted the portfolio to contain the fewest number of ETFs suitable for the strategy execution while providing reasonable diversification. Twelve turned out to be the sweet spot.

Portfolio allocations - The portfolio weightings and other data are given in the table below. Four funds have double allocation. The assumption is we are putting $2,000 each in the first four ETFs and $1,000 in each of the other eight, for a total starting portfolio value of $16,000.

Portfolio volatility - Ten of the twelve components of the portfolio (exceptions are VHT and GXC) have had a higher 3-year average beta than the index SPY, albeit only modestly. However, this is in fact a desirable factor for the implementation of the strategy, as it becomes clear later.

Vanguard dominance - Seven of the twelve ETFs ended up being Vanguard funds. I didn't set out with a proclivity for Vanguard funds, and in every category I compared all the worthy choices before deciding. The chosen funds just happened to be the best in their sectors in my opinion. Additionally, even though none of the selections was based on the lowest expense ratio, it's a nice plus that most Vanguard ETFs have the lowest cost in their sector.

The 12-ETF Immortal Portfolio and SPY

Data as of 11/01/2011

Symbol

$ Price

Shares
Held

$ Market
Value

%
Weight

% Total Return
12 Month

%Total Return
3 Year

% Total Return
5 Year

RSP

45.36

44.09

1,999.92

12.50

4.98

16.60

1.37

VB

68.43

29.23

2,000.21

12.50

5.25

15.34

2.20

VEA

31.97

62.56

2,000.04

12.50

-8.40

7.49

n/a

VWO

40.61

49.25

2,000.04

12.50

-12.78

20.36

5.27

VDE

98.91

10.11

999.94

6.25

14.85

11.42

5.45

VGT

61.89

16.16

1,000.14

6.25

5.10

18.93

4.53

VHT

58.80

17.01

1,000.19

6.25

7.72

10.45

2.48

VNQ

56.27

17.77

999.92

6.25

5.96

16.16

-1.05

GXC

64.47

15.51

999.93

6.25

-18.70

19.80

n/a

EPI

19.53

51.20

999.94

6.25

-28.06

20.61

n/a

ILF

44.15

22.65

1,000.00

6.25

-14.14

21.56

10.17

GNR

50.72

19.72

1,000.20

6.25

-5.44

n/a

n/a

12-ETF

16,000.47

100.00

-3.41

15.90

3.57

SPY

122.00

131.15

16,000.56

100.00

4.90

10.31

-0.21

RSP (Rydex S&P 500 Equal Weight)

This ETF contains the same S&P 500 stocks, however they are all weighted equally, unlike the S&P 500 Index (SPX, or its tracking ETF, SPY,) which are market cap weighted. For example, as of the end of September 2011 the top 10 weighted stocks accounted for 20.58% of the index.

Although RSP is on average slightly more volatile than the index, it has outperformed it over 1-year, 3-year, 5-year, and since its inception on April 24, 2003.

VB (Vanguard Small Cap ETF)

VB tracks the performance of the MSCI US Small Cap 1750 Index. Like RSP, VB is more volatile than the SPX index, but has beaten it over 1-year, 3-year, 5-year, and since its inception on Jan 26, 2004. While researching small cap ETFs, I considered several other funds in the space, including IJR, IJT, SLY, and several others. VB has had one of the best long-term returns, one of the lowest costs, and is one of the broadest.

VEA (Vanguard MSCI EAFE ETF)

The MSCI EAFE Index is made up of approximately 970 stocks of large-cap companies located in 22 countries in Europe, Australia, Asia, and the Far East (Asia). The portfolio must have substantial exposure to the rest of the globe, and I believe this ETF is the best for our objective. A clear alternative to VEA would have been iShares MSCI EAFE Index Fund (NYSEARCA:EFA), and in fact the latter has outperformed slightly in 1- and 3-year periods, although VEA has outperformed year-to-date. In most other metrics they are similar, except EFA has six times the size and ten times the volume. But I think with its lower expense ratio (0.12% vs. 0.35%) VEA will do better in the long run.

VWO (Vanguard MSCI Emerging Markets ETF)

The MSCI Emerging Markets Index includes approximately 748 stocks of companies located in emerging markets around the world. As noted earlier, the emerging markets are perhaps the most significant long-term theme, so they play a major role in the portfolio. The clear alternative to VWO would have been iShares MSCI Emerging Markets Index (NYSEARCA:EEM). I chose VWO because it has outperformed EEM over 3- and 5-year periods and also has a lower expense ratio (0.22% vs. 0.69%), lower turnover (11% vs. 17%), and the number of its holdings is slightly higher (901 vs. 871).

VDE (Vanguard Energy ETF)

VDE tracks the performance of the MSCI U.S. Investable Market Energy 25/50 Index, an index made up of approximately 168 all cap U.S. companies within the energy sector. I considered other energy ETFs, including XLE and IXC. The XLE (Energy Select Sector SPDR) lost out since it holds only 44 stocks vs. VDE's 168. The reason I prefer more holdings is that the fund will have to encompass some smaller companies in addition to the giants. As for IXC (iShares S&P Global Energy,) I would have gone with it due to its global index, however, VDE, since its inception on Sep 23, 2004, has consistently outperformed it. Furthermore, exposure to foreign energy companies including BP, TOT, RDS.B, PTR, and PBR is already achieved through VEA, GXC, and ILF.

VGT (Vanguard Information Technology ETF)

VGT tracks the MSCI US Investable Market Information Technology 25/50 Index. The obvious alternative to VGT would have been XLK (Technology Select Sector SPDR). I picked VGT for two main reasons: It has outperformed XLK over 1-year, 3-year, 5-year, and since its inception on Jan 26, 2004; and secondly, it holds 409 stocks vs. XLK's 83.

VHT (Vanguard Health Care ETF)

VHT tracks the MSCI US Investable Market Health Care 25/50 Index. The alternative ETF would have been XLV (Health Care Select Sector SPDR). I picked the Vanguard fund over its Spider counterpart for the same two reasons as picking VGT over XLK. It has outperformed XLV over 1-year, 3-year, 5-year, and since its inception on Jan 26, 2004; and secondly, it holds 293 stocks vs. XLV's 53.

VNQ (Vanguard REIT ETF)

VNQ tracks the performance of the MSCI US REIT Index. Any broadly diversified portfolio must include some REITs, and in my opinion this is the best ETF in the space.

GXC (S&P China BMI index)

Picking the right China fund for the portfolio was the toughest and took some extra research. Here are the most relevant China ETFs:

Data on Select China ETFs

Total Return as of 10/31/2011

Symbol

1 Year

3 Year

5 Year

# of Holdings

$mil Fund Size

FXI

-18.16%

48.19%

36.46%

27

6,862

GXC

-16.81%

70.17%

n/a

184

729

PGJ

-19.57%

53.06%

37.06%

185

294

YAO

-17.66%

n/a

n/a

192

63

MCHI

n/a

n/a

n/a

150

52

FCHI

-18.27%

55.78%

n/a

115

40

PEK

-18.84%

n/a

n/a

300

17

FCA

n/a

n/a

n/a

51

3

After a thorough comparison I decided on GXC. The close alternative would have been MCHI. I excluded the small-cap only (NYSEARCA:HAO), the Honk Kong focused (NYSEARCA:EWH), and the A-share focused ETFs (NYSEARCA:PEK) and CEFs (NYSE:CAF) and considered all the selection factors mentioned earlier.

EPI (WisdomTree India Earnings Fund)

There are several single-country funds dedicated to India to choose from. I excluded the CEFs (IFN, IIF), as well as INP, which is an ETN. I picked EPI over the other major ETFs (PIN, INDY) because of its underlying index, whose methodology gives a larger allocation to small and mid cap stocks than the other India ETFs. Large-cap stocks still get the largest allocation in EPI, but companies with market capitalization of under $10 billion make up approximately 40% of the total holdings. Furthermore, both PIN and INDY hold only 52 stocks to EPI's 142.

ILF (iShares S&P Latin America 40 Index)

This fund provides exposure to the several emerging market countries in Latin America. Of the three broad ETFs covering the region: ILF, GML, and FLN, I chose the best known and the largest. But GML was a close 2nd choice. In the process, I excluded the small-cap and the single-country ETFs and considered all factors discussed earlier.

GNR (SPDR S&P Global Natural Resources)

The commodity/natural resources fund was the one that I had the hardest time picking. I wanted a single broad fund that covered both hard and soft commodities. After delving into the sector's various types of funds, including the futures-based, the physically backed, and the equity-based funds, I decided on the following:

  • No ETNs (such as UCI, GSC, DJP, GSP).
  • No futures-based ETFs (such as DBC, GSG, GCC, USCI) due to the well known futures markets contango issue. Note that the commodity ETNs are also futures-based, albeit in an indirect way. I should also point out that the tax treatment of futures-based ETFs is different from equity-based ETFs. These ETFs file partnership tax returns instead of corporation returns. Accordingly, tax information is provided to shareholders on Schedule K-1 rather than Form 1099, which can cause some inconvenience or even tax complications.
  • Physically-backed ETFs (such as GLD for gold) are impractical (and non-existent) for literally all commodities except for precious metals due to low value-to-weight ratios.

After eliminating the above the only ETF type left is the equity-based ETFs. Among the few ETFs in this category are GNR, HAP, and IGE. I picked GNR over HAP mainly due to the tracked index, and over IGE due to its global nature vs. IGE's U.S. focus.

Now we have chosen our 12 ETFs and ready to execute.

Strategy Execution

Now we can embark on purchasing the ETFs to build the portfolio. It's best to accumulate the shares over several months rather than all at once due to market unknowns. We buy a portion when the market, or a particular ETF gets hit. Once all ETF allocations are in the portfolio, the annual rebalance clock will start. We hold the portfolio for one full year from the day the portfolio was fully in place, and then perform our annual rebalances.

Procedure for the First Annual Rebalance

The rebalance anniversary date for the portfolio will be November 1st of every year. I will track and rebalance the portfolio annually. I should point out that the term rebalance is often used to refer to trades that would bring the various allocations to predetermined levels, but that is not the usage here. On or around the anniversary date we sort the 12 ETFs by their 12-month total return percentages. The top three (#1-3) and the bottom three (#10-12) performers will be our six candidates for trading. Generally we will be selling a portion of the top three and putting the proceeds into shares of the bottom three.

We make trades according to the following rules:

  • Sell 30% of the shares in #1 and use the proceeds to buy shares in #12
  • Do the same for (#2 and #11) and (#3 and #10)
  • If any of the double allocation ETFs (RSP, VB, VEA, VWO) are among the top three, you only sell 15% of the shares (half of 30%); that is, they are always treated as if they had equal weight as the other eight.
  • If the difference in percentage returns in any of the above three pairs is less than 10%, do not make the trade. We will do fewer trades this year!

Procedure for the Second Annual Rebalance and Beyond

After the first year, we basically repeat the process once a year, on or around the rebalance anniversary. The rules are the above plus the following:

  • If an ETF has made it to the top three for the second year in a row, sell 40% of the remaining shares, or if it's made it for the third year in a row, sell 50% of the remaining shares. If a double allocation fund is involved, the share sales would be half (20% and 25%).
  • For any ETF, if three out of the last four trades (regardless of the elapsed time,) have been in the same direction (three buys or three sells,) do not trade this ETF in the same direction.
  • If an ETF's weight has gone above 20% of the portfolio total, do not add to it.
  • If an ETF's weight has gone below 5% of the portfolio total, do not sell from it.
  • If an ETF lands among the six trade candidates but it is restricted from trade because of any of the above rules, then substitute the next ETF in line.

The reason for the progressive sales of 30%, 40%, 50% when an ETF outperforms in consecutive years is that the odds of continuing to outperform diminish with time. The concept of reversion to mean is significantly more applicable to ETFs than to individual stocks.

Reasoning Behind the Strategy

As I mentioned before, I've formed the strategy based on three basic concepts: buy and hold, buy low, sell high, and stay in the market at all times. I've heard persuasive arguments about how far ahead an investor would have been if the investor was out of the market in 2008 and early 2009 during the credit crisis. True, but how many investors got out of the market at the right time? How many of those who correctly got out got back in the market in March 2009 to catch the ensuing huge bull market? Most of those who got out in time, stayed out or worse, stayed short well past the March '09 bottom, thus canceling out any benefits they got from their correct initial move. So, it is as damaging to be out of the market during a bull run as it is to be in the market during a bear, probably worse. Since it's impossible to predict market's gyrations, it's better to be invested at all times and fine-tuning the portfolio periodically than to jump in and out of the market with big bets.

Let me illustrate why I believe the system will have a good shot at beating the index in the long term. To do so, I make an assumption at the outset: If the 12-ETF portfolio is simply bought and held (no trades,) its performance on average over several years will roughly match that of the S&P 500 Index ETF SPY. This is a reasonable assumption considering the portfolio has so far actually beaten the index over 3 and 5 year periods. For the sake of illustration we further assume that SPY's constituents are our 12 ETFs. To beat SPY, all we need to do is wait until it's low to buy it, and wait until it's high to sell it. Then we stay out until it's low again. The trouble is obviously we don't know when those occasions are.

Since the index is the average of all the moves by its constituents, and the constituents move up and down at different rates and times, instead of trying to capture the highs and lows of the whole portfolio as a unit (such as getting in and out of SPY), the strategy allows us to captures the tops and bottoms indirectly in a relative, incremental, and piecemeal manner. Every time we rebalance by selling a portion of a top mover and adding to a laggard, it's like selling a piece of the index at its relative high and buying it right back at its relative low. So, we won't try to predict the real tops and bottoms and we won't need to. We let the market do it for us over time. Furthermore, we won't be jumping in and out of the market since our incremental sell high, buy low happen at the same time! Clearly, in the short run most of our rebalances probably won't capture the major tops and bottoms. However, the key is that market volatility, the concept of mean reversion, and time are on our side. Eventually some of our trades will capture the major highs/lows and the cumulative effect is how the portfolio can get propelled ahead of the index.

Source: The Immortal Portfolio: A 12-ETF Strategy To Beat The Market