My permanent portfolio with individual stocks

[ngcpa]

On 9-21 posted about the PP I planned to start:

http://gyroscopicinvesting.com/forum/ht ... ic.php?t=2

Well the CD’s matured and I did start it.  I actually bought an equal amount each of 14 stocks.  I bought individual stocks instead of a total market index like VTI, because I don’t see VTI as a suitable substitute for what Harry Browne describes in his book “Why the Best-Laid Investment Plans Usually Go Wrong”?. He states on page 250: the most appropriate stock-market investments for a PP are: stocks that normally are highly volatile, mutual funds that invest in volatile stocks and warrants.  He even provides some lists of volatile stocks (page 336) and mutual funds (page 340).  I don’t see VTI as a suitable substitute. 
      I feel the key is to find the right stocks.  I think I found a way to do this.  I understand that a big part of the success of the PP is that the three big moving parts (stocks, long term treasuries and gold) are highly uncorrelated, and the rebalancing captures a lot of value.  I have collected data for many stocks, ETF’s and mutual funds.  For each data stream I have the price at the end of each qtr and at the close the day before a dividend is paid.  For each data stream I calculate annual gains starting at any qtr (other than the last 4), the % of qtrs that had gains for the next year, the average gain (of those years that had gains) and the average loss (for those years that had losses).  I also print this data in tabular form with other measurements I won’t go into here.  The following is a small part of this table:
                      % of      avg    avg
Symbol    qtrs  gains      loss    gain
--------  ----  -----    -----    ------     
VUSTX      100      85      6.5    10.0
Spot gold  151      59      11.2    22.9
VTSMX      78      77      18.7    18.2   
PRPFX      115      82      5.2    10.0
PP X 4        73      90      2.5      9.5

Most of you know what these symbols represent, but just in case:
VUSTX is Vanguard’s open-ended long-term treasury fund.  Not the greatest representative, but the most data easily available.
VTSMX is Vanguard’s open-ended total stock market index. 
PRPFX is the PP mutual fund.
PP X 4 is an investment of 25% each of VMFXX (Vanguard’s federal money market fund), 25% VUSTX, 25% spot gold and 25% VTSMX.  I wrote a program that simulates
investing in data streams with chosen percentages and chosen rebalancing methods.  For this run I used the standard (15% & 35%) rebalancing bands (or 40%).
        The important columns are the average losses and average gains.  Notice that the gold, long term treasuries and even both the PP’s  have bigger average gains than losses.  So in a year period that has a gain, it is generally twice as big as losses sustained (on average).  Notice the stocks (VTSMX) has average losses as big as gains.  Fortunately stocks (like all the others) have more gains than losses.  It seems to me, that if I could find a group of stocks that have larger average gains than losses (like the other components), the results would improve dramatically.
      Here is the same table with the 14 stocks I have chosen :

                      % of    avg    avg
Symbol    qtrs  gains    loss    gain
----        ---    ----    ---    ------ 
ARLP      49        80    10.6    50.2
AZO        82        72    11.4    34.6
BCPC      83        80    24.1    45.0
BLL          88        69    12.3    28.6
CHRW      56        81      4.1    32.4
FAST      87        81    16,2    35.7
KMP        77        86      5.7    30.5
NKE      104        76    19.8    42.2
ROL        95        82    10.2    23.1
ROST      51        81    13.4    31.0
RYN        71        79    13.5    24.6
SJI        51        87      3.7    21.0
SKT        73        78    10.5    24.9
SRCL      61        81      8.6    43.4

It should be noted that both ARLP and KMP are limited partnerships.  If you hold them in a taxable account you will get a K-1 and have fun preparing your taxes.

I feel this is a nicely diversified group of companies:  coal, auto parts, chemicals, packaging, transportation, fasteners, pipelines, running apparel, pest control, clothing, timber REIT, gas distributor, outlet REIT, and environmental..  Unfortunately there is not a lot of history for some of these companies (ARL:P & SRCL), and some of the companies have a long history, but I couldn’t find data for more than 51 Qtrs (ROST & SJI). 
When I ran my simulation program with these stocks instead of VTSMX along with the other 3 components (spot gold, VMFXX, & VUSTX) over 49 qtrs.the average annual gain improved about 5%. 
I found when I ran my simulation program without VMFXX but the just the moving parts, I found the average annual gain went up another 2%, although it was a slightly more rocky ride. 
I reran everything for 18 years (replacing about ½ the stocks with others that I felt were good that I had more data for.  I had similar results.  Since I am planning to have the PP for a long time, I am willing to have a slightly bumpier ride for the added return.  Therefore, I have a 3 x 33.3% PP!
Well, I am ready for the bricks, barbs and hand grenades.  Fire away.
Norm

[AdamA]

Since I am planning to have the PP for a long time, I am willing to have a slightly bumpier ride for the added return.  Therefore, I have a 3 x 33.3% PP!

Do you have any cash set aside in case of an emergency?  It would stink to have to sell out of your bonds, stocks or gold during a downturn b/c you suddenly needed some cash.

As for the stocks...I'm not a fan of stock picking, but as long you can keep you're expenses low, it probably won't hurt much.  The important thing is that you're comfortable with it. 

Keep us updated. 

[ngcpa]

Previously I was self employed and now retired, so I always have (need) a fair amount of cash around in Vanguard's GNMA for example.  I also have other investments (a lot of perferreds, some CD's, stocks, etc.)  I also get social security.  I am not betting the farm on this, although it is  a significant amount of my net worth.
Norm

[stone]

Norm, are you going to rebalance each stock as an individual entity? I mean if just one of your stocks falls or rises, will you sell some gold or bonds so as to rebalance that individual stock?
I wittered on about this a bit in:
http://gyroscopicinvesting.com/forum/ht ... ic.php?t=7

About not holding cash:- in 1981 (I think), cash performed well and the other three plumeted. Basically if the Fed hikes up short term rates much above inflation then they induce that scenario. Harry Browne called that condition "tight money". A cash free 3x33% will suffer badly then.

[ngcpa]

Stone:
I will not rebalance stocks as individual entities.  Over its history (about 18 years) VMFXX has had an average annual return of about 3% and some years even paid about 6%.  Recently it hasn't paid anything.  I certainly don't see the fed hiking short term rates above inflation any time soon.  Even if they did, it doesn't mean the other 3 will plumet.  According to Craig:

https://web.archive.org/web/20160324133 ... /#more-299

for 1981:
Year TSM ST Bonds LT Bonds Gold Returns
1981 -3.7 18.9 1.9 -32.8 -3.9

So without short term bonds the total return would have been -11.57% instead of -3.9%.  I could live with that.  Also I do not see short term bonds returning 18.9%.  I remember long term interest rates (for mortgages) were about 14% then and the 30-year treasury was about 12% then.  I don't see that happening anytime soon either.  My time horizon is different than a lot of others being 66.  Another minor advantage of having  short-term treasuries is rebalancing is triggered every 5 (according to my back testing) years instead of 3-4 years (using 40% bands).
Norm

[stone]

Norm, I'm interested because I'm in the odd position that my better half wants to keep most of our savings as cash that she takes care of. I deal with the remainder. I do have some of what I look after as UK index linked savings certificates (sort of cash) but that makes up less than 25% of what I'm looking after so in terms of rebalancing I'm almost a 3x33% since the cash my better half looks after is not part of the same pool for rebalancing. In your back tests, has your hypothetical 3x33% held up pretty well?

[ngcpa]

My backtests went very well, but I have only gone back 18 years.  Whether you go 4 X 25% or 3 X 33%, I have found using "the right" individual stocks to be far superior to the total stock market index.  By "the right" stocks, I mean stocks that go up 2-3 times as often as they go down, and have average gains 2-3 times as large as average losses.  In the data Craig used (that I linked to), I don't see how short term treasuries had a CAGR of 7.5% and about 1/4 of the years had double digit gains.  I am not questioning Craig, as it is not his data.  I wonder what this data consists of.  I plan to eventually have about 25 different stocks in my PP.
Norm

[stone]

Norm, good to hear that the 3x33% did OK in your backtests. About choosing the best individual stocks- my problem with that is that all we can know for sure is what was good not what is going to be good. Did you mean that you did backtests where you pretended to choose stocks in say 1990 purely on the basis of what they did up to 1990 and found that that prior performance WAS predictive of what happened in 1991 (essentially a momentum strategy)?

[moda0306]

Yes... I'd also like to know how to pick these individual stocks.

It seems to me that if the best fund managers out there can barely do it reliably, with all the assets at there disposal (computers, experience, very smart employees, etc) how in the world am I going to do a better job?

Further, unless you have a big enough portfolio to justify carrying a large enough number of stocks to avoid the huge risks of under-diversifying within the stock market, then you're left with huge trading charges to boot.  Even if you have $100k to trade, with 20 stocks that's $5k per stock on average, and you'll still be subject to 5% losses if any one of the company's go under.

And it doesn't take much for that to happen.  Think of Arthur Anderson, Enron's accounting firm.  The vast majority of those accountants were probably engaged in honest behavior (or at least weren't doing anything related to Enron) and then Enron came along and sunk the entire accounting firm with them.  Absolutely insane that one client could bring down an entire firm, but it happened.

I loved netflix a couple years back, and probably would have made it one of my "picks" if I was into individual stock picking, but look at them now.  A few marketing mistakes and they're in huge trouble.

So I don't know what the "magic number" is, but it most definitely would have to be at least 20 stocks to get you some diversification, and to be honest I think many more would be appropriate.  Managing this would be a headache, as well as probably futile in your attempt to beat the market in the long-term.

And we truly are talking about a free lunch here.  By diversifying to 500 or 5,000 stocks, you avoid the multitude of risks that are engrained in any one individual stock, but still get the average likely return of all stocks.  Unless you can spot value that will later be appreciated by the market as a whole, you're not going to beat this situation very well.

[stone]

Thanks Clive, I think my policy of filling my ILSC quota but otherwise being 3x33% doesn't seem too foolhardy.

[ngcpa]

Stone and moda0306:
        I pick stocks for my PP based on the criteria I mentioned - they have more years of gains than losses and the average gain is bigger than the average loss.  I am shooting for at least 2:1 in in both cases.  For the most part I didn't have to look very far to find these stocks, as I have previously owned all but 3 of them (FAST, NKE & RYN).  In fact when I started my PP I transferred 7 of them from my Roth IRA at Fidelity to a new Roth IRA I set up at Fidelity (I want to keep my PP holdings separate).  A few others I sold in my taxable account at Fidelity and then rebought them for the Roth  PP.  I still own some other stocks that I like (mostly utilities and other pipelines) .  I didn't use any of the utilities because they don't fit the PP criteria I established and they aren't volatile (which is one of the main reasons why I bought them).  Since I already chose KMP, I didn't want anymore pipeline companies in my PP.  For the most part I have chosen stocks based on previous history.  Although this might not work well for mutual funds, I think it does for stocks.  I have found that if a stock has performed well for 10 years, generally it will continue to perform well.  You do have to watch them as they will go bad for many reasons.  The most common is mistakes made by maagement, such as expanding into a new area they don't know enough about or borrowing excessively.  I am quite familiar with all the stocks in my PP and have watched them and owned most over a long period of time.  Some of them I wouldn't normally own as they are far too volatile for my taste, but fit the PP criteria well.  One example is ARLP.  This stock has gone up consistently and even pays a large dividend (slightly over 5%), but it is an incredible rollar coaster ride.  Hope that helps.
Norm

[stone]

Norm, I'm also interested to hear why you decided not to trim/add to your individual stock holdings so as to harvest their individual volatility. Unlike you I have never owned an individual stock but in principle I find it hard to understand why a PP with perhaps 10 individual, decent, diverse stocks each held at 2.5% wouldn't behave better than the 25% total stock market set up. To my understanding, the advantage would predominately come from capturing the individual volatility (eg one stock goes up whilst another goes down).

[ngcpa]

Stone:
      I don't have my simulation program set up to do that currently, although I could make a copy and change it to accomplish individual stock rebalancing easy enough and compare the results.  I have a feeling using 40% bands on the individual stocks might cause a lot more rebalancing than I would want (most of these stocks double every 3-4 years - which hopefully will continue).  My inclination is to just go with the 14 stocks and rebalance them only when I do a rebalance based on the 3 main categoties breaking out of the 40% bands.  I would be more inclined to replace a stock that has blown up and otherwise just let them run.  I do plan to add to my PP in one year when more CD's come due.  I will probably add 6-11 stocks.  20-25 stocks total would be nice, as then any one stock would represent a little more than 1% of the entire PP.
Norm

[stone]

Norm, whenever you add to or reduce your stock allocation overall you could do so in such a way as to get the individual stocks ballanced with repsect to each other. In principle if you only ever added as a totally new stock and only ever sold stocks that had gone badly sour, then you would never realize any of the capital gains?? From what I can see, companies are like trees, they grow then they wither and die or get knocked over. Any strategy has to be based around harvesting the growth while they are growing. Nowadays dividends seem less and less of an effective way to harvest the profits of many stocks.

[ngcpa]

Stone:
    Without having to modify my program code, I was easily able to make 2 runs.  I used all 14 stocks without anything else (gold or treasuries of any kind) and had data for the past 49 quarters.  For the first I didn't rebalance at all and the second I used 40% bands.  Some interesting results:
$ 10,000 grew to $ 112,304 in a little over 12 years without rebalancing.  With rebalancing it grew to $ 119,243 (a 6% improvement),  but rebalanced 13 times.  I also tried rebalancing with other bands (20%, 30% and 50%) as well as rebalancing once a year.  Using the 40% bands had the best result, although all were fairly close.  It would be easy to determine when an individual stock is above the 40% as it would be above 10% of the total stocks (1.4 x 1/14 works out real nice).  At the low end a stock would have to fall below roughly 4.25% of the total stocks (.6 X 1/14).  Originally I was going to buy 12 stocks at 3K each and have 36 x 4 or $ 144K.  However after I decided to go with only 3 components, I thought 14 stocks at 5K each would be best and now have 3 x 70 or $ 210,000.  1/2 of 1 % means it would grow an additional $ 1,050 a year (my best guess), so you have convinced me it would be worth watching the individual stocks and rebalancing them as necessary.  I am a little concerned with all my modeling and hope it all holds up.  Someone once said the trouble with simulation is that it is a lot like masturbation - if you do it enough you could begin to think it is the real thing.
Norm

[stone]

Norm that's great. I have to make it doublely clear that I have only thought this would in principle be a good idea but in practice have my stock allocation as three diverse closed end funds. Personally I would be inclined to keep the number of stocks limited to those that you can thoroughly research and keep a close eye on. Look at how few stocks make up the vast bulk of Warren Buffet's portfolio. There are quarterly arbitary gyrations in the price and rebalancing those is what provides your profits. There are also situations where a company is about to die. You need to know the companies well enough to pull the plug as soon as the fundamentals tell you it is no longer viable. Of course index funds also hold dying companies. You can pick and choose. I really hope it works out well. May the force be with you!

[stone]

Norm, I also suspect that it might make quite a difference if the rebalancing is within the context of a PP rather than within just the stocks. If when ever you rebalance out of LTT or gold into stocks, you end up adding more to the lagging stocks and less to the leading ones, then that could avoid most stock to stock rebalances (or if you are retired and drawing down, then you sell more of the leading stocks).

[stone]

Clive, "Its typically better for funds than stocks however, as it in effect ploughs increasingly more the lower the price declines and as such is anticipating a mean-reversion/recovery."

Clive if you look at charts of individual stock prices (which I know you do) there is a tremendous amount of price gyration (ie volatility available for capture as profit). Your example where you posted up charts of various Dow stocks that had wildly different one year charts really emphasized that. Do you not think that it is reasonable to expect recovery of stocks that you have researched well and understand what they do and how they do it and can be seen to be doing a decent job? Obviously you would need to avoid certain stocks that do business that you don't understand and to pull the plug/not rebalance repeatedly into them if they made a potentially terminal screw up. 

[tndiehard]

It seems to me that the risk of individual stocks is too high.

In this day of Zillions of mutual funds and ETF's it seems that you could find something with a Beta and Volatility that tickles your fancy.

But then I am a bit conservative and like the PP because it seems to smooth the Volatile Stock Ride

In any case if you are comfortable with it, kudos and Good Luck

Gerry

[moda0306]

I've been caught, and heavily so, by individual stocks in the past that appeared to have all the right qualities that then suddenly turned heavily southwards due to some really silly slip up by the board/director.

I've since opted to never invest heavily into any one stock/investment and spread/diversify much more widely so that no single failure becomes uncomfortable.

What seems good value can become even better value. What seems like a ceiling can become a floor. Diversifying and reacting to take profits and cost average into decliners (when it seems reasonable to do so) is a lot easier than trying to out-research the millions of other researchers who are all attempting to do the same.

Clive, whatever you do don't delete this post.

The amount of risks you avoid by diversifying into a group of stocks is amazing, and like you said, all it takes is one bad slip up or scandal.  Look at Netflix for example.  You avoid these risks by diversifying into an index, but what you give up is an opportunity to do what is near-impossible anyway... pick a stock that is underpriced AND will soon be properly priced.

One thing I think people need to understand is that the market can be mispriced for a LONG time, and it can get worse before it gets better.  Do you really want to expose your self to the whim of one ego (the CEO) while waiting for that price to bump up a bit?

Further, what good is buying a stock underpriced by 10% if the whole market's going to blow up due to a financial crisis you didn't see coming because you were too focused on ratios and quarterly earnings to see the forest through the trees.

[stone]

moda, whilst I think it is important to diversify I still haven't shaken off my index fund skeptism. If you had 2.5% of your portfolio as a single stock, that still seems to me less of a concentration risk than holding 25% gold or 25% LTT is. To my mind holding say a FTSE 100 index in 2007 would have given a concentration risk towards banks. Holding ten seperate stocks at 2.5% of the portfolio each with only one of them being a bank would to my mind have constituted less of a concentration risk.

[moda0306]

2.5% in single stocks would expose your stock portion of the portfolio to still a moderate amount of risk.

If during a crisis, just two of your stocks went belly-up (not unreasonable to imagine), that's a 20% loss without even considering the massive drop in price of my other stocks.

That said, you may have a point looking at what sectors make up an index fund... what portion of the FTSE 100 is financial stocks?

[stone]

moda I think the proportion made up by the banks went through a total hiatus in 2008. The two biggest banks, RBS and HBOS, dissolved into nothing and the other banks also shrank down massively. I think the proportion to banks is now not too high but it certainly was too high in 2007. I just googled and got a motley fool article about it but I don't want to register to get the quotes .

Imagine you had gone into the tech bubble in the late nineties with only one or two tech stocks and had rebalanced out of them. You would have been keeping or perhaps even building your holdings for coca cola, exxon mobil etc. By contrast, as you rebalanced out of a stock index fund you would have been throwing out the non-bubble stocks along with the bubble bathwater.

[moda0306]

I think they should have "everything but" index funds... this would be especially useful for owners of closely held businesses that want to diversify outside of there industry.

For instance, "S&P 500 w/o Financials" would be appealing to a lot of people.

[ngcpa]

I don't understand some of the comments on this thread.  I have said I am using 14 stocks in my stock section of my PP.  I do plan to increase the number to 20 or 25 next year.  Since I also am not using cash, stocks (like  gold and LTT) represent  1/3 of my overall PP.  So currently each stock does represent about 2.5% (actually 2.38%) of my total PP.  If a stock were to drop 20% and I determined future prospects didn't look good, I would replace it.  If two stocks were to go bad, it wouldn't mean the PP would drop 5%, but more likely about half of that at most.  Companies generally don't go bankrupt over night, and I doubt they would be worth 0.  More importantly, VTSMX (Vanguard's total stock market index) once had a one-year loss of 38%.  According to my calculations, the odds are almost 1 in 4 that VTSMX will have a loss when held for a year.  Furthermore the average loss (for those years with a loss) was almost 19%.  Is this any better?  I would be very surprised if my 14 stocks didn't out perform VTSMX.
Norm