Howdy, Stranger!

It looks like you're new here. If you want to get involved, click one of these buttons!

In this Discussion

Here's a statement of the obvious: The opinions expressed here are those of the participants, not those of the Mutual Fund Observer. We cannot vouch for the accuracy or appropriateness of any of it, though we do encourage civility and good humor.

    Support MFO

  • Donate through PayPal

An Almost Never Lose Money Portfolio (P: MJG)

MJG
edited May 2011 in Fund Discussions
Hi Guys,

I have noticed a very risk averse thread running through several Mutual Fund Observers (MFO) postings. Wealth preservation is a paramount objective.

In the spirit of that observation, I recall century old adages for wealth preservation from the Bible and from the house of Rothschild. Dividing the portfolio into three equal pockets that enhance diversification seemed to be the strategy of the day, over multiple centuries. Never place all your resources into the keel of a single ship was good advice yesteryear, and is still solid advice today.

I also remember that Richard Young, publisher of the Intelligence Report financial newsletter, has advocated a wealth preservation strategy for decades that has embodied a 50/50 mix of balanced mutual funds.

Along that line of reasoning, I have constructed a simple 3-unit, equally funded portfolio of two balanced mutual funds and the diverse Permanent Portfolio mutual fund (PRPFX). The two mutual funds that I selected for the portfolio are the Dodge and Cox (DODBX) and the Vanguard Wellesley (VWINX) mutual funds. These funds were chosen as candidate actively managed funds with cost containment considerations. The expense ratios for PRPFX, DODBX, and VWINF are 0.77, 0.57, and 0.28, respectively. That’s an average annual cost of 0.527; not bad for actively managed portfolios.

Of course, Index funds and ETFs also offer similar cost containment elements and are attractive alternate candidates to serve the same wealth conservation goal.

A 15-year timeframe was selected for the purposes of this brief study to manage time and to include the turmoil associated with two market downturns. Data was collected from the Yahoo business and the Paul Merriman websites for the 1996 to 2010 timeframe.

The data were entered into the StatView statistical computer code. Analyses were completed for the original data sets and for both a 50/50 split of equity/bond passively managed proxies and for a 30/30/30 mix of the identified actively managed mutual funds.

The results of the study are summarized at the end of this posting and are extracted from the StatView analyses.

Note how the conservative funds have delivered returns that are similar, and for this limited timeframe, sometimes superior to the S&P equity standard. Because correlation coefficients are usually well below the perfect correlation level, diversified portfolios can be assembled that reduces volatility without compromising returns performance.

For those investors who can not tolerate negative annual returns for whatever reason, the lower volatility levels (that’s a lower standard deviation) translates into fewer negative years over an extended holding period.

For the 15-year timeframe examined, the S&P 500 produced negative results 4 times. That’s a about 27 % of the time and is relatively consistent with the historical record of roughly 30 % downward annual performance.

In contrast, the individual component portfolios and their 30/30/30 mix delivered far fewer disappointing returns. In particular, the actively managed 30/30/30 mix generated below zero rewards only 1 time. The passively directed 50/50 Index portfolio only reduced negative results to 3 times.

Here are some of the statistical findings:
Correlation Coefficients
Portfolio Mean Return Std. Dev. DODBX VWINX PRPFX S&P 500 Bonds 50/50 30/30/30
DODBX 9.73 14.40 1.00 Index Mix
VWINX 8.01 7.65 0.77 1.00
PRPFX 6.13 7.65 0.57 0.48 1.00
S&P 500 8.55 20.36 0.81 0.58 0.47 1.00
Bonds 6.88 8.23 - 0.13 0.05 -0.17 -0.25 1.00
50/50 Index 6.89 9.36 0.79 0.66 0.44 0.98 -0.20 1.00
30/30/30 Mix 7.94 8.76 0.95 0.86 0.75 0.76 -0.11 0.76 1.00

As Albert Einstein famously observed "Anyone who has never made a mistake has never tried anything new.", and "Learn from yesterday, live for today, hope for tomorrow. The important thing is to not stop questioning."

Good luck to all you dedicated investors, especially to those who face the daunting challenge of avoiding all negative returns.

Best Regards,

MJG


Comments


  • ============ Correlation Coefficients ============
    Mean Std. S&P 50/50 30/30/30
    Portfolio Return Dev. DODBX VWINX PRPFX 500 Bonds Index Mix
    ------------ ------ ------ ----- ------ ----- ----- ----- ----- ------
    DODBX 9.73 14.40 1.00
    VWINX 8.01 7.65 0.77 1.00
    PRPFX 6.13 7.65 0.57 0.48 1.00
    S&P 500 8.55 20.36 0.81 0.58 0.47 1.00
    Bonds 6.88 8.23 -0.13 0.05 -0.17 -0.25 1.00
    50/50 Index 6.89 9.36 0.79 0.66 0.44 0.98 -0.20 1.00
    30/30/30 Mix 7.94 8.76 0.95 0.86 0.75 0.76 -0.11 0.76 1.00

  • Hi Investor,

    Thank you for correctly formating my garbled table. I was not alert to the fact that the website would compress my inputs.

    How did you accomplish this task?
  • I would suggest adding GTAA in lieu of the Dodge & Cox fund otherwise fully endorse the approach. However would not buy PRPFX and instead adopt Harry Browne's true 4 x 25 portfolio with equal parts of SHY, TLT, VTI or SPY and GLD
  • I formatted your text in Notepad (or your favorite text editor which uses fixed with fonts) and placed it between the following HTML tags.

    <pre>
    .... your pre-formatted text goes here ....
    </pre>
  • I wouldn't have guessed that DODBX would have such a high correlation with the S&P 500. Very interesting exercise MJG and a good suggestion for a simple portfolio.

    I just don't know if there are enough elements for me to tweak with though;)
    I think foreign stock & bonds, mid cap, and small cap are underrepresented here.
  • Hi Anonymous,

    Thank you for your thoughtful and excellent recommendations. It is difficult to take issue with any of them.

    My proposed multi-balanced mutual fund portfolio was simply cobbled together to illustrate the concept. In no way did I mean to imply that this provisional portfolio was anywhere near optimum. I mostly liked it because of its low cost character. Other alternate selections are always welcomed.

    As a sidebar, I was very pleased to note your reference to Harry Browne. I knew him both as a twice presidential candidate for the Libertarian Party and as a fund manager. I miss him in both arenas.

    Although I am not a Libertarian, I do endorse some of their approaches to control government spending and especially liked Browne’s writings on monetary crisis and personal liberty.

    On the investment front, I had the great privilege of meeting him several times at the Las Vegas MoneyShow. He was always a soft-spoken, patient, gentleman.

    I mostly remember him for his careful presentations that were delivered without visual aids and almost without any notes. His entire talk was captured by him on a single 3 X 5, tattered index card that seemed to be densely annotated, including along the cards edges.

    He passed away much too early in 2006.
  • MJG, an excellent exercise. I have done similar things to put together a "core" group of balanced and allocation type funds and used return and volatility (stdev) to compare the groupings. I never looked at correlation between funds. I like that view.

    A question, if not too much trouble, could you show the yearly returns for the 50:50 and 30/30/30 portfolios? It looks like even at 2 sigma, which I like to use for forecasting the range of expected returns, the probability of the 30/30/30 mix loosing more then 10% in any given year would be pretty low with the 3 arbitrary funds you used.
  • Hi MikeM.

    No problem. Here are the annual returns for te two groupings.

    Year Index 50/50 3-Part Mix
    2010 9.9 14.1
    2009 13.9 21.2
    2008 -15.0 -17.7
    2007 7.0 6.6
    2006 8.9 13.0
    2005 2.9 6.6
    2004 7.0 8.6
    2003 15.6 13.7
    2002 -5.0 2.0
    2001 -2.5 7.0
    2000 0.9 10.5
    1999 9.6 3.5
    1998 17.9 5.6
    1997 19.8 16.4
    1996 12.5 8.0
  • MJG,

    You need to do the formatting on Notepad (assuming using Windows) and paste here between the tags. The default message editor here is using a variable with font so it is hard to align things correctly while composing message in the box.

    Prepare your tables on your host computer and copy to clipboard and paste it into text box and don't reformat after pasting. It will appear correct (or you can try preview tab to check)

  • MJG
    I am still questioning in my mind how DODBX has such a high correlation with SP 500
    where one is a mixture of equity and bonds and the latter is pure equity ?
    Please comment.

    Burt S.
  • edited May 2011
    Great discussion Investor, MJG and all. Sorry I don't really understand the charts, but DODBX being one of my largest holdings for nearly10 years I'll attempt to add what I can. Do occassionaly read their reports. I bought DODBX as a sedate "balanced" fund but quickly realized it was something more aggressive than might appear. Next to PRWCX and OAKBX which I own it is much more volatile both short and long term. Daily it will be 50% more volatile on average than these two tamer cousins. Compared to the S&P it appears to have near 70-75% of the volatility.

    Why? D&C has long been bearish on bonds (incorrectly in recent years) and maintains a 65-70% weighting to equities in DODBX while the two comparitive funds I mentioned tend to run closer to 60% equities. Also, I suspect a somewhat higher weighting to technology, HP long being one of their biggest holdings. Comcast is another big one. D&C uses essentially their DODIX income fund as the income component in DODBX. While DODIX is a relatively short positioned stable fund with a high yield component, it doesn't react in the same way a long term bond component would to dampen equity volatility. Keep in mind also that Dodge and Cox includes foreign holdings in their domestic DODGX which in turn translates into some foreign exposure in DODBX. Without looking it up, foreign equities are probably in the vicinity of 10-15% of DODBX's holdings, which of course means additional volatility compared to the S&P.

    Just some thoughts. Hope I havent muddied discussion too much or stated anything too incorrect regards DODBX.


  • Dodbx -33 % 2008
    Oakbx -16
    jabax -13
    If I were choosing a fund I would choose the lower of the 3 above provided that the Sortino ratio confirmed the above .
    Also VWINX Vanguard Wellsely seems to be too over weighted in bonds to call balanced in my opinion.
    So the question is what is the perfect pair of balanced funds to add to permanent PRPFX
    for the program? Would it work better to to have 4 funds added to PRPFX. 2 pure eqhuity and 2 pure bonds.
    Burt S.
  • Hi prinx,

    Hank posted a far more insightful reply than I can; his DODBX knowledge is more comprehensive than mine.

    I understand that although DODBX probably has a 60/40 target asset allocation, it has had a higher percentage of equity holdings for a long time now (north of 70 %). In fact, that is one of my reasons for selecting them as one of my two candidate Balanced funds.

    DODBX represented the aggressive Balanced fund holding in my sample portfolio. VWINX represented my conservative sample mainly because of its stated policy to hold only a 30 % equity position. The two distinctive entities seemed like a reasonable strategy to mitigate portfolio risk. The PRPFX holding was added to further mitigate risk with its holdings in gold/silver and Swiss francs.

    Also DODBX tends to invest n rather Large Cap and somewhat value-oriented corporations. Therefore, it should always exhibit a somewhat tight correlation to the S&P 500 index.

    So, I guess I was not surprised by the 0.81 correlation coefficient between the S&P 500 and DODBX. Keep in mind that a correlation coefficient of 0.81 translates into an R-squared value of 0.65. That means that about 65 % of the DODBX annual returns can be explained by (or coupled to) the Index's yearly movement. Since DODBX typically owns double-digit levels of foreign positions, I am inclined to accept the correlation as an acceptable measure of their investment choices over the last decade and one-half.

    Thank you for your interest. It is always necessary to push away from an analysis and mentally challenge the results to test if it satisfies the commonsense criteria.
  • edited May 2011
    Bert: Tried to link a M* X-Ray of DODBX which seems to confirm MJG's and my estimations. Over 70% equities / 17.5% in tech which is more than the S&P / 12% foreign stocks.

    Link not working well. Just follow it to xray homepage and enter DODBX and a value of $100. It will break everything down for you. Kinda neat.

    Unfortunately am not qualified to answer your question, having little knowledge of VWINX or JABAX.

    The biggest knock against DODBX is that is a huge fund and no doubt suffers from some of the problems of asset bloat. The large cap concentration MJG mentioned is no accident. Large funds gravitate to large cap stocks. If they tried buying or selling smaller issues they'd rock the markets too much and drive the price of a company up while trying to buy more shares or down while trying to sell.

    Perhaps the nicest thing about it is that the .53 ER is low for a managed fund. Some will argue ER doesn't matter. I think over a very very long time (20-30 years) it does matter.

    http://portfolio.morningstar.com/NewPort/Free/InstantXRayDEntry.aspx?entrynum=10&runMode=MSTAR
  • Yes ER matters but only if your net return is well below the net return after the ER is subtracted and especially if you are a pure buy - hold investor. I like MJG's concept. It would be scarry for most investment managers to find out that this is probably superior results with less risk than the majority of the practioners can obtain. Just think if one could tweak it to provide the best 3 making up the 30% buckets or rather the 2 buckets that could go along with the PPRFX bucket. This could be the find of the past generation. MJG or some of the other great posters may have the tools to determine this. YEs........... this could be a sensation.

    Burt S.
  • I'm not a fan of Janus overall, but I think JABAX is a fine balanced fund. Additionally, the Perkins funds remain solid offerings out of Janus.
  • I would like to know how to set up a study to determine the the best pair to match with PRPFX. to do a back study going back 15 -20 years to see what would happen if the the 3 30% buckets would do through various market cycles or evaluate them by Sortino Ratios

    Burt s.
  • All,
    I would not recommend including two balanced funds in a portfolio such as this as there is too much correlation between Wellington/Wellesley and the Dodge & Cox fund.

    Again, my recommendation is to try and vary your strategies by selecting one balanced fund (ie Wellington), Harry Browne's Permanent Portfolio or PRPFX if you want a tilt more towards inflation and like paying a higher ER by purchasing a fund and then for the third part I tend to favor Mebane Faber's trend following strategy and would use his ETF GTAA even thought the ER is on the high side.

    For the more creative you may want to consider the Merger (or AQR Arbitrage Fund) and Gateway Fund. All the other active funds are noise.
    Heather
  • PRPFX didn't do so hot in the 1990's. Gold didn't appreciate much in the 80's and 90's.
  • I think you are missing the point. This is not to find a great fund but a combination that can withstand the volatility and risks through the market or business cycles over a period of years and still have a fair return to take advantage of the magic of compounding. The question I have is about how to back track your choices to see what would Have happened had you chosen a given combination at a given point years ago.
    Burt S.
  • Exactly, I think some has missed the point that this exercise was not a recommendation on particular funds but a strategy that will not shoot the lights but produce somewhat more reliable acceptable returns by taking advantage imperfect correlation between 3 balanced funds.
  • Ping: MJG
    Would appreciate your comments on some of the dialogue in the past few days.
    I think your original thoughts are intriguing and I wonder if you can expand on my comments and those of others.
    Burt s.
Sign In or Register to comment.