Bogleheads

[macclary]

MT good points about TIPS. In my mind the only time it makes sense to own tips are as a un-taxed institution that has liabilities tied to the actual CPI figures. If someone recommends TIPS to you as a private individual run away

Regarding going back to a gold standard I actually do think it is possible because it happened in this country 20 years ago. Greenspan decided to put us on a virtual gold standard in 1987 by raising the Fed funds rate when gold climbed in nominal price, and lowering the rate when gold dropped. http://www.goldworld.com/image....enspan.gif As you can see we left that virtual standard around '97 I guess because Greenspan wanted to use monetary policy to lean into the internet bubble.

If a country wanted to go from fiat to a gold standard the keys would be to do it quietly and not publicly commit to the standard too early. It may make sense to never commit to what the contents of your currency basket are so that the currency traders never know if they have you against a wall or not.

[macclary]

I received a private message asking about how to design PP like portfolios. I decided that this might be of interest to others so I'll bring it out here even though it might get me excommunicated ;->

In my opinion Harry Browne discovered a family of portfolios not just one portfolio allocation. The first characteristic of this family of portfolios is that they have strategic protection against loss in the 4 economic climates. The second characteristic is that these portfolios have small draw-downs when you look at historical performance both in sample and out of sample.

In order to find portfolios in the PP family we should only look at allocations that have sufficiently large allocations to gold, long bonds, and stocks to survive the respective economic conditions. These portfolios should also have back-tests that look fairly smooth over a period that includes examples of the 4 economic climates. If there are major losses in recent history then that is a good clue that we are on the wrong trail. The one thing that it impossible to do immediately is to achieve the same true out of sample reputation that the 4x25 have rightfully earned since HB first suggested the PP. Don't take this point lightly: "updated modifications" to the 4x25 would need to have real money put into them for decades to be as well tested as the canonical PP. Perhaps there are compelling reasons why someone would be willing to pursue this testing for a different allocation in the PP family.

Base Line: equal parts long bonds, cash, TSM, and gold returned about 9.71% compounded 1972-2008. The worst year was 1981 @ -4.28% (using the data I have on hand).

Scenario 1: someone who only owns cash and CDs, would like to start investing but the prospect of loosing any money - even a few percent in a year - keeps them from committing. In this case it is possible to look for a portfolio that has the key characteristics of 4x25, but is lower in volatility and return.

http://www.riskcog.com/portfol....k5p8bzf5pc

Portfolio Allocation: 42.2% TBILL, 20.0% GOLD, 20.0% LTGB, 12.1% Intl Value, 3.6% 5 Yr T, 2.1% LCV
Compound return = 8.71%
Worst year: 1994 0.75%

To construct this portfolio I set minimums for gold and long bonds arbitrarily at 20%. I then added a variety of equity and fixed income asset classes to the mix and found which combination had the least bad worst year with a target of return of 8.71%. (8.71% is 1% less per year than the 4x25.) This portfolio happens to have not had a losing year since the 1970s but it certainly could have a losing year in the future. Due to the large amounts of cash and bonds I think many people would agree that this portfolio is likely, but not guaranteed, to have smaller nominal losses than the traditional 4x25 in the future.

Scenario 2: someone who is used to investing in a more aggressive style such as 70/30 stock/bond portfolio finds the PP very appealing for its ability to survive inflation or deflation but they can't bring themselves to dial back the return they hope to earn to the more "bunker" like historic returns of the PP.

http://www.riskcog.com/portfol....d1il17e135

Portfolio Allocation: 36.7% LTGB, 21.9% GOLD, 14.8% EM, 6.7% Intl Value, 6.6% 5 Yr T, 5.3% Pacific, 4.2% ST Trsry, 3.8% SCV
Compound return = 11.72%
Worst year: 2008 -3.39%

To construct this portfolio I chose a target return two percent higher per year compared to the 4x25. I then added some additional asset classes and found an allocation without large losses in the last few decades.

Scenario 3: the person from scenario 2 decides they don't want to track 8 funds, and they only want to put 5% of their portfolio into emerging markets. Oh, and they don't want all of those decimal allocations...

http://www.riskcog.com/portfol....81fi20e9e5

Portfolio Allocation: 35% LTGB, 33% SCV, 20% GOLD, 7% ST Trsry, 5% EM
Compound return = 11.73%
Worst year: 1990 -5.84%

These PP like portfolios all have significant commitments to stocks, gold and long bonds. They also have (hypothetically) ridden through almost 4 decades of turbulence without flying apart. Now all we need to know is if this approach will meet expectations for the next 4 decades...

[MediumTex]

macclary, what you are talking about is the same thing that HB talked about in his earlier writings where he recommended many different PP-like allocations based upon the macroeconomic environment you felt was unfolding.

The evolution of HB's thinking is interesting to study.
_________________
"A Permanent Portfolio should let you watch the evening news or read investment publications in total serenity."
-Harry Browne

[Rose21]

Since the Permanent Portfolio is ultimately premised on correlations (at least with respect to economic conditions, if not asset classes inter se), is there a point at which we should reevaluate whether something fundamental has changed that should alter the approach?

I'm speaking here primarily of treasury bonds and the fact that prices/yields are not being driven by the market, but rather by purchases by the Fed, and that by all appearances this kind of price manipulation cannot be sustained over the long term. More broadly, I'm referring to the massive amount of leverage in the market that, when it unwinds, may well cause every asset class that can't be picked up and moved to plummet in tandem.

Yesterday and today provide a glimpse into what has been concerning me about the PP for a while now. The bond market is not reacting. Everything is being sold off.

[Kevin K]

Thanks, MT, for your very direct and clear "rant" about TIPS. Why would one trust the same entity that does everything it can to understate the actual rate of inflation to keep our income sufficient to deal with it in times of need?

[Lbill]

Kevin - I dunno. I'm not sure I trust nominal treasuries any more than TIPS treasuries. Only difference is how expected inflation is accounted for. Right now, long treasuries include only a small inflation premium in the interest rate. But that's as misleading as the Fed's CPI figures, if not more so. I think one of the main drivers of low treasury rates right now is that the U.S. has decided to copy the Japanese model and lower rates to zero. So, now we have the "carry trade" on our shores, instead of in Japan. Unless we have the mother of all depressions, this will lead to a very bad fate for long treasuries and stocks eventually.
_________________
"Whenever you find yourself on the side of the majority, it is time to pause and reflect." ~ Mark Twain
"A foole and his money is soone parted." - J. Bridges, 1587

[MediumTex]

[MediumTex]

Just a note about inflation, and I think that this is something I have touched on in prior posts, but it bears repeating:

Without rising wages, real inflation in an economy cannot exist.

So for anyone who is scared of inflation, don't watch price levels (because they can send false signals, as in 2008), watch wage levels.

It is when wages begin rising in tandem with prices that you have a problem on your hands. Absent increasing wages, all rising prices mean is looming demand destruction and the validation of the Austrian economics view of the world (again, see 2008).
_________________
"A Permanent Portfolio should let you watch the evening news or read investment publications in total serenity."
-Harry Browne

[Clive]

The following data, although not very presentable, is in a better format to cut and paste than had I uploaded a image of the data.

The data was constructed by taking each of PP's component part yearly gains and subtracting inflation (CPI) from that value in each case. Having identified the after inflation gain for each year the annual average from each year starting from 1972 up to 2008 was then calculated together with the standard deviation.

Using the annual after inflation gain and and stdev an approximation of the compound average (annualised) was then made and which the following data shows.

So in effect we can scan down to any start year and see what the after inflation CAGR up to 2008 might have been.

The overall average of 3.8% is perhaps then an indicator of what average income might have been withdrawn whilst still allowing the fund to pace inflation over time. At the individual years level that income is reasonably stable when averaged across all four components, however at the individual component level there are some pretty wild variations around that average.

As an example consider 1999, from that year to the end of 2008 stocks averaged poorly relative to inflation (-3.56% after inflation growth (loss), however LT Bonds at 5.81% and and Gold at 8.74% compensated for the poorer stock performance and collectively across the PP four component set a 3.22% after inflation growth was achieved up to 2008.

Had a PP been started in 1983 then stocks would have done well, averaging 6.69% after inflation gain up to the end of 2008, LT bonds would also have done well (7.19%) whilst gold came in at a -0.97% below inflation average gain up to 2008. The overall collective average was 4.13% for 1983 to 2008.

Just to clarify, the figures in effect assume that yearly rebalances of the portfolio back to 25% equal weightings were made.

Year start to 2008 end year TSM LT Bonds ST Bonds Gold Average
1972 4.64 4.12 2.97 2.56 3.57
1973 4.40 4.17 3.03 1.58 3.29
1974 5.36 4.52 3.12 0.24 3.31
1975 6.96 4.87 3.27 -1.04 3.52
1976 6.34 5.02 3.41 0.09 3.72
1977 5.92 4.83 3.42 0.41 3.65
1978 6.51 5.25 3.63 -0.10 3.82
1979 6.74 5.75 3.83 -0.96 3.84
1980 6.57 6.45 4.00 -2.56 3.61
1981 6.14 7.42 4.12 -2.56 3.78
1982 6.98 8.06 3.96 -0.70 4.57
1983 6.69 7.19 3.61 -0.97 4.13
1984 6.22 7.60 3.54 -0.24 4.28
1985 6.49 7.45 3.34 0.91 4.55
1986 5.62 6.67 3.08 0.81 4.04
1987 5.23 6.00 2.77 -0.02 3.50
1988 5.59 6.65 2.79 -0.75 3.57
1989 5.19 6.72 2.85 0.31 3.77
1990 4.27 6.40 2.79 0.68 3.54
1991 5.23 6.74 2.75 1.17 3.97
1992 3.91 6.37 2.54 2.19 3.75
1993 3.73 6.50 2.46 2.95 3.91
1994 3.46 6.02 2.40 2.20 3.52
1995 3.92 7.24 2.80 2.72 4.17
1996 1.99 5.86 2.33 3.67 3.46
1997 0.73 6.76 2.40 4.67 3.64
1998 -1.48 6.32 2.24 7.72 3.70
1999 -3.56 5.81 1.89 8.74 3.22
2000 -6.02 7.87 2.14 10.06 3.51
2001 -4.99 6.85 1.74 12.70 4.07
2002 -3.66 7.64 1.28 14.74 5.00
2003 -0.05 6.43 0.46 13.44 5.07
2004 -4.97 7.67 0.52 12.66 3.97
2005 -8.54 8.44 1.08 15.26 4.06
2006 -12.49 10.16 1.99 15.34 3.75
2007 -23.61 16.61 2.70 12.73 2.11

Average 1.82 6.79 2.70 3.91 3.80

I should perhaps mention that whilst stocks average is relatively low, much of that is a draw-down arising from the more recent large stock price declines.

This next chart summarises that data and gives a feel for how that overall 3.8% after inflation average gain was formed over time.

[Clive]

Here's similar data, but for the UK

Year start to 2008 end year Shares Gilts Cash Gold Average
1972 4.18 3.31 1.11 2.66 2.82
1973 4.05 3.76 1.13 1.60 2.64
1974 5.61 4.51 1.18 0.41 2.93
1975 8.78 5.90 1.47 -0.72 3.86
1976 7.30 5.72 1.97 0.61 3.90
1977 8.00 5.96 2.18 0.66 4.20
1978 7.21 5.21 2.30 0.86 3.89
1979 7.45 5.63 2.34 0.47 3.97
1980 7.95 6.34 2.60 -1.20 3.92
1981 7.54 6.36 2.70 -1.46 3.79
1982 7.77 7.04 2.77 -0.44 4.28
1983 7.22 5.84 2.62 -1.23 3.61
1984 6.62 5.66 2.55 -0.94 3.47
1985 5.85 5.80 2.43 -0.93 3.29
1986 5.48 5.82 2.32 0.27 3.47
1987 4.72 5.76 2.11 -0.37 3.06
1988 4.71 5.44 1.93 -0.12 2.99
1989 4.70 5.58 1.96 0.86 3.28
1990 3.62 5.99 1.90 0.78 3.07
1991 5.04 6.56 1.83 2.99 4.11
1992 4.41 6.11 1.66 3.55 3.93
1993 3.65 5.53 1.34 3.03 3.39
1994 2.34 4.27 1.28 2.32 2.55
1995 3.19 5.73 1.32 3.15 3.35
1996 2.00 5.00 1.36 3.40 2.94
1997 1.09 4.98 1.47 5.17 3.18
1998 -0.48 4.06 1.65 8.14 3.34
1999 -1.58 2.43 1.39 9.42 2.92
2000 -3.90 3.33 1.18 10.63 2.81
2001 -3.28 2.97 1.00 11.61 3.08
2002 -1.66 3.31 0.58 13.02 3.81
2003 2.96 2.72 0.60 14.14 5.10
2004 0.23 3.52 0.62 15.32 4.92
2005 -2.02 3.46 0.60 21.31 5.83
2006 -8.37 2.52 0.23 19.95 3.58
2007 -17.39 6.26 0.35 28.29 4.38

Average 2.92 4.96 1.61 4.92 3.60

[helmut]

[MediumTex]

[MediumTex]

For those who think they have a pretty good idea about what the future holds, I would like you to read the following piece written by Ron Paul:

[Maestro G]

Hi Medium Tex,

I think you make very interesting points and a strong argument against the dependence on TIPS as a hedge against significant inflation. It all makes much sense.

What I have a hard time keeping in perspective and excepting, relative to (and actually because of) your clearly learned and well thought out view, is the very contrary view by any number of extremely intelligent, experienced, financially astute investment professionals for whom TIPS are a cornerstone and Gold a speculative pariah of their consistent allocation advice and overall investment philosophy.

I think of Zvi Bodie in particular, David Swensen (at least for non-institutional investors), Bogle, Bill Bernstein, Swedroe, etc.. Certainly, these learned men, have all considered the concerns you raise regarding TIPS, and have an excellent grasp of the historical behavior of Gold in different economies. Why then, do you think, (and this is not a challenge, I ask all rhetorically) they reach such very different conclusions about the value of these two asset classes within an investment portfolio?

Is it just an entrenched investment philosophy

Maestro G

[macclary]

I think that congress would be the ones out of a job if they passed $100 minimum wage ;>

Good question Maestro, I think that it requires some armchair psychological analysis to try to understand why people believe different things. From a psychological standpoint maybe these men bought gold in the late '70s and then lost money for decades until they swore off of it when selling for a loss.

Another psychological take is that it is a much easier task to talk most people into buying something with the warm snuggly name "Treasury Inflation Protected Security". Each of the 4 words packs positive emotional impact. One investment advisor told me that they call TIPS the "feel good option", if their client was contemplating doing something really stupid then they would bring up TIPS because they knew they could talk them into that instead!!

If you want your investing commentary to be widely accepted by individuals and advisors TIPS have a lot going for them. There are exceptions but selling people on gold is a challenge regardless of how good the theory and historical performance may be.

Last point: for the vast majority of companies and institutions TIPS could be fine. This is because no matter what they buy these entities are probably not going to survive a major disruption in which the government is unable to make good on the TIPS commitment. People and families will out-live the vast majority of companies and institutions and actually need to plan differently.

Each sector of the PP requires a different mental state. Stocks require you to think only about the go-go years and what will do well when industry starts making money hand over fist. Gold on the other hand requires a more morbid outlook where you consider what you would sell if you needed to escape a country falling into civil war or equivalent. In the last century the German people lost all of their liquid wealth twice - at least if they only ever thought about investment with the happy hat on.

[craigr]

[MCSquared]

[craigr]

[craigr]

[MediumTex]

[craigr]

[MediumTex]

[SquawkIdent]

Interesting stuff...

http://www.koamtv.com/Global/story.asp?S=11355575

[meckaneck]

Fellow Bogleheads,
I was not sure where to post this question however currently my family is contemplating adopting the PP versus a 50/50 stock bond portfolio. My question to the PP followers: If someone is in a wealth accumulation mode (40 years old) and saving for retirement does it make sense to adopt Harry Browne's PP? My concern is that a few of the asset classes have no real rate of return after inflation and since we have 20-25 years until retirement would it be wiser to adopt a 50/50 boglehead portfolio or still consider the PP?
Thanks in advance for any help.

[macclary]

Hi meckaneck, there is actually a whole other thread dedicated to airing the differences between the 50/50 and the PP that might be interesting to you. If you are considering a buy/hold/rebalance AA type portfolio then one thing you should keep in mind is that the individual risks and returns of each component are only important in context of the whole portfolio. Here is a pertinent quote from Swedroe:

[snowman9000]

Here is a very good article that explains why you might want gold, the drawbacks, and answers some of the criticisms and concerns. Specifically, why gold has no intrinsic value.

http://www.lewrockwell.com/north/north772.html

[MediumTex]

[mathu1968]

[mathu1968]

[macclary]

Money is valuable idea because it allows humans to form markets and best benefit from the skills and resources others can bring to market. Gold is useful because its total supply is relatively stable, and its scarce enough that you only need to carry a little bit around. It is also easily formed but resistant to corrosion. US dollars are very handy too because they are honored by a great number of people, and there is a useful electronic funds transfer system in place.

Here is an interesting WSJ story about how canned mackerel acts as currency in prisons!!

"Mackerel Economics in Prison Leads to Appreciation for Oily Fillets"
Packs of Fish Catch On as Currency, Former Inmates Say

http://online.wsj.com/article/....96481.html

[MediumTex]

[Clive]

-- deleted --

[Skbosma]

First time poster! Many thanks to the great contributions made in this thread and others on Bogleheads.

I'm planning to move my own version of the PP which will be as follows:
50% Stock (25% US; 25% Intl)
25% Bonds
12.5% Gold (implemented through an ETF)
12.5% Cash

Several contributors have shown charts for historical performance of various portfolios. Since I'm a novice at this I was hoping someone could provide annual return for the above portfolio since 1972 (?), I remember seeing that as a starting point for the HB PP analysis.

Many thanks to whoever can help.

SK[/list]

[Lbill]

Sk - can you specify the exact bond and cash holdings you have in mind?
_________________
"Whenever you find yourself on the side of the majority, it is time to pause and reflect." ~ Mark Twain
"A foole and his money is soone parted." - J. Bridges, 1587

[Skbosma]

LBill, let's assume the bond and cash holdings are as suggested by the standard HB PP. Thanks for your help!

[Lbill]

Using Simba's spreadsheet the returns from 1972-2008 are the following:

TSM...........25%
EAFE..........25%
LTT............25%
STT...........12.5%
Gold..........12.5%
CAGR.........9.75%
SD.............9.87%
Sharpe.......0.44
Best Yr.......29.6% (1985)
Worst Yr....(-14.9%) (2008)
--------------------------------------------------------
TSM...........25%
Tot Int.......25%
LTT............25%
STT...........12.5%
Gold..........12.5%
CAGR.........10.05%
SD...............9.98%
Sharpe.........0.47
Best Yr.......29.8% (1985)
Worst Yr....(-15.5%) (2008)
------------------------------------------------------------------
TSM...........25%
LTT............25%
STT............25%
Gold...........25%
CAGR.........9.62%
SD.............8.49%
Sharpe.......0.48
Best Yr.......41.9% (1979)
Worst Yr....(-4.1)% (1981)

STT - 2 yr. treasuries
LTT - long term treasuries
EAFE - developed international
Tot Int - developed + emerging

The difference between the first and second ports is the use of EAFE vs. Total International. It appears that adding a little Emerging Market (with TI) improves things a bit.

The 3rd portfolio is the classic PP. As you can see, the classic PP limits downside risk compared to the other two porfolios. For example, it lost 0.7% in 2008 compared to around 15% for the other two ports.
_________________
"Whenever you find yourself on the side of the majority, it is time to pause and reflect." ~ Mark Twain
"A foole and his money is soone parted." - J. Bridges, 1587

[macclary]

[Skbosma]

Thanks Lbill and Macclary!

I cannot bring myself to do 50% in Gold/Cash and only 25% in Equities. The higher risk is no problem - I was 80/20 (stocks/bonds) when the market collapsed last year and didn't lose any sleep over it!

I happened to find the HB PP discussion a few days ago and really like the elegance/simplicity of the portfolio working in different economic environments.

Appreciate all the contributions on the forum both for and against the PP.

[Lbill]

Here's an interesting one. 30% in stocks divided between Small Cap Value and Emerging Market. 60% in 5 year treasuries. 10% in Gold. As you can see, from 1972-2008 it outperformed the Classic PP on all measures except the best year gain, with lower risk. It only lost 2.2% in 2008. BTW, 60% in 5-year is a near substitute for 30% LTT + 30% STT. So this is essentially a 30/30/30/10 portfolio. The reason it outperforms is that the returns from SCV and Small or Emerging Market international have been much better than TSM, and the reduction of Gold to 10% mitigates the underperformance of Gold from 1982-2002, in exchange for foregoing some of the larger gains from gold during the 1972-1981 period.

SCV...........25%
EM...............5%
5 YR ...........60%
Gold............10%
CAGR.........10.7%
SD...............7.3%
Sharpe........0.69
Best Yr........25.6% (1982)
Worst Yr.....(-3.6%) (1994)
2008..........(-2.2%)
_________________
"Whenever you find yourself on the side of the majority, it is time to pause and reflect." ~ Mark Twain
"A foole and his money is soone parted." - J. Bridges, 1587

[macclary]

A few pages ago there was another question about the idea of using intermediate bonds instead of HB's prescribed short and long combo. In happy circumstances there is fairly similar performance because the duration is similar. What you are giving up however is convexity, liquidity, and income during great depression II. Keep in mind that a deflationary depression could last decades after your 5yr treasuries have reset to 0% interest.

For these solid reasons I will stick with HB's barbell bond portfolio!

I don't think there is any gain at all from the intermediate term bond approach. If your 401k or equivalent plan does not offer long bonds, and buying long bonds in another account is not practical, then try the 50% 5yr idea. 40% 5yr and 10% cash could also work and would give you some more liquidity.

[MediumTex]

[Tramper Al]

[macclary]

I just saw this today, its a new ETF called "CPI" that aims to

"
provid[e] a “real return” or a return above the rate of inflation, as represented by the Consumer Price Index,Q
"

http://www.indexiq.com/indexes....index.html

How do the "experts" at financial institutions create real return? Well cash, gold, and long bonds of course! Notice no TIPS, intermediate treasuries, corporate bonds, zeros, high yield bonds or any other detritus that is brought up occasionally. (There are Yen and Oil funds further down in the holdings at less than 1%)

Top 4 holdings:

ISHARES BARCLAYS SHORT TREASURY BOND FUND 54.10 %
SPDR BARCLAYS CAPITAL 1-3 MTH T-BILL ETF 28.96 %
ISHARES BARCLAYS 20+ YEAR TR 8.04 %
SPDR GOLD TRUST 7.38 %

There is no equity exposure since this is basically a low return, hedged replacement for nominal bonds.

[Clive]

[macclary]

Thanks, those are good suggestions Clive. There are few features available that start to address some of what you want to see. You can view the portfolio returns by rolling decades. Also you can search portfolios based on returns for median decade or worst decade.

For analyzing the risk of a portfolio I think that it is important to look at the full data set. For analyzing returns I think it makes good sense to look at the minimum and maximum returns over various sub-periods as you suggest.

[Clive]

I should clarify that I'm not picking on PP as I do actually hold a PP (as one part of the whole). Instead I've tried to highlight the risks that some may overlooked.

As an example an investor who started a PP in say the 1980's might have seen real (after inflation) returns of around 3% p.a. up to the mid 1990's, whereas index tracker investors might have seen 10% p.a. real returns. Such 7% relatively lower returns would (and did) put many off PP.

Similar situations exist for many other investment blends, for example emerging markets made good gains and as such including EM's as part of a blend based on 'cheap' start date levels may attract investors to include EM's as part of the set held. However in the forward (out of sample) time direction those investments often prove to be the poor performers.

Makes you wonder if seeking out historic blends that have produced the lower rewards might be the better choice for forward time holdings

[macclary]

Exactly! In my research I have found that historic returns are not predictive of future returns This seems obvious, but basically all of finance is built on the opposite hypothesis. MPT for example seeks the "highest return per unit risk", in practice this sets you up to get your head shot off out of sample! A better approach is "lowest risk for a given return", this approach is much better out of sample: the safest portfolio in a previous period does seem to be among the safest portfolios going forward.

I coded up and tested a few different "highest return" type optimizers for my web site, but they were much too dangerous! I did decide to include an optimizer after some deliberation but it is the second type: "safest portfolio for a given level of return". Some people may be confused about how these are different, but if you do the type of multi-period testing Clive is talking about then you will see the difference immediately.

Buying the portfolio with the lowest historical return is like switching from a momentum system to a value system. There is a paper that shows that instead of choosing between value and momentum, you actually get outstanding returns by doing half of each because they are inversely correlated! http://papers.ssrn.com/sol3/pa....id=1363476

[Clive]

-- deleted --

[MediumTex]

[macclary]