HBPP compared to other strategies on The Retire Early Homepage

[moda0306]

Where the PP potentially 'fails' - depending upon your definition of failure - is in the average rewards it provides. If you're content to accept an inflation pacing type total net return then the PP appears to more often than not fulfil that objective.

Clive,

I just don't see where you think 10% average annual return is merely matching inflation over the past 10 years.  If that's the gauge, then stocks themselves have barely beat inflation as well.

I know I could scour this board for where you're getting your "real return" figures but it seems to me that I'm not quite understanding or agreeing with your calculation of inflation over the past 40 years. 

Was a large portion of those U.S. 1930's losses silver?

Thanks again for all the useful input and infotainment.

[MediumTex]

The thing is gold's performance isn't all that unreasonable when you consider its nature as not a tracker of CPI but a gauge of scary things to come, mostly involving currencies and purchasing power.

Saying that gold is related to the CPI is sort of like saying that the severity of a thunderstorm is related to the pattern on the weatherman's tie.   

[MediumTex]

Clive and moda,

I thought we all agreed that the PP's historical returns have been along the lines of inflation + about 4%.

[moda0306]

MT,

Yeah that sounds about right... I am pretty sure if I wanted to juice my PP returns with a VP I'd probably add some foreign & EM stocks and some kind of small cap fund, maybe throw in some energy stocks... along with some of Clive's Relative Strength stuff but with other volatile PP assets as my hedge instead of cash... Just off hand without thinking about it much.

So I think Clive has some absolutely great input here, but I 1) don't trust what seem like repetative technical trends nearly as much as relationships based on fundamental macroeconomic principals (RSPP vs HBPP), and 2) think that we may have some disagreement on inflation.

Thanks Clive.

[MCSquared]

Clive and moda,

I thought we all agreed that the PP's historical returns have been along the lines of inflation + about 4%.

MT, Simba's dataset reflects inflation + 4.5% (or so) since 1972 for the HB PP.  Note that this is a US investor and it is not an exact replication as the Vanguard Long Bond data does not match the duration of the Long Bond piece of the PP (as outlined by HB).

[craigr]

Clive and moda,

I thought we all agreed that the PP's historical returns have been along the lines of inflation + about 4%.

The portfolio has returned in the +3-5% real after inflation range from my data as well. Harry Browne observed and stated the same thing.

The portfolio has not just matched inflation. In fact, I can't find a rolling 10 year period where it ever was zero or negative real returns at all. This is a very unique feature of the portfolio compared to its contemporaries. It has managed to pull off a real rate of return in that +3-5% range.

-- Craig

[gizmo_rat]

Where the PP potentially 'fails' - depending upon your definition of failure - is in the average rewards it provides.

What got me thinking was that the original article does a good job of summarising and comparing the strategies available to a fairly average investor (with or without advisor assistance).

Failure would therefore be defined as choosing the strategy that failed to support a 4% annual withdrawal + inflation. On the face of it most strategies available to the average investor can't do that, which is shocking.

I am pretty sure if I wanted to juice my PP returns with a VP I'd probably add some foreign & EM stocks and some kind of small cap fund...

Which was going to be my next question, but I'll save it for another time / thread.

Thank you both.

[craigr]

The portfolio has returned in the +3-5% real after inflation range from my data as well. Harry Browne observed and stated the same thing.

T Bills (3 month duration) are sometimes considered as the risk-free rate.

Since 1980 (according to Simba's spreadsheet), T Bills have averaged 1.89% real.

Generally I'd expect T-Bills to pace inflation over the mid to longer term.

Relative to the risk-free rate therefore, any PP comparisons should deduct 1.89% from the 'real' PP figure for a fairer longer term average indication of the PP's potential.

3% to 5% declines to 1.1% to 3.1% (average 2.1%).

1980 is a peculiar year to start a comparison for T-Bills. A lot of turmoil going on. A look at the period of 1972-1982 shows a return of -0.38% for T-Bills during bad inflation. From 1972-2010 a rate of +1.07% is shown. This is about what I'd expect long term from T-Bills.

The portfolio has not just matched inflation. In fact, I can't find a rolling 10 year period where it ever was zero or negative real returns at all. This is a very unique feature of the portfolio compared to its contemporaries. It has managed to pull off a real rate of return in that +3-5% range.

Its not unique. Coffee House hasn't had a negative 5 year real return, which if you consider REIT and International to be commodity like, the Coffee House is a form of 40% stock, 20% commodity, 40% bond type blend.

REITs were not commonly available even 20 years ago. The ones that did exist were quite expensive. There was no investable REIT index going back through the entire data set people use. Same thing with Small Cap Value funds. The data probably is loaded with survivorship bias as well. So the Coffeehouse Portfolio is built on some assumptions which may be good, or may not.

But even if we assume the data is good, the portfolio was about 40% more volatile over the past 40 years. The past 10 years the Coffeehouse was more than twice as volatile as the Permanent Portfolio with slightly less return. In 2008 the portfolio also saw a -20% loss. In the years 1973-1974 the Coffeehouse portfolio experienced a -10% CAGR with the real losses being -18.5% after inflation was factored in. In those years the Permanent Portfolio saw real gains of +3.48% CAGR.

30% TSM, 20% 5 year T ladder
30% Commodity, 20% IBonds

also hasn't had a negative 5 year real. During rising rates (inflation) generally Commodity and IBonds do OK. During declining rates TSM and 5 Yr T do OK. Which is a 60-40 speculative/safe type blend where stocks and commodities have low/no correlation.

I-Bonds have not existed during a period of high inflation so no telling how they will do. Commodities will do better, but I suspect gold still wins.

The PP's comparable 1.65% real but with lower down years is a bit of an illusion as the PP also periodically can encounter -20% drawdowns, but in the US and UK that hasn't aligned to year end dates so it seems that they've not occurred when based on year end yearly reviews.

Dates are arbitrary, I agree. Markets are emotional creatures. My portfolio had about a -12-15% drawdown max in 2008, but that only lasted very briefly (less than a month, more like a week??? I don't recall). The drawdown then has lasted a much shorter time IMO than other approaches. Only fairly recently are stock/bond portfolios recovered/almost recovered to the prior 2007 highs. So you're talking about losses that have taken years to recover from vs. the Permanent Portfolio which recovered relatively quickly from these kinds of market shocks or wasn't affected at all.

[AgAuMoney]

gold was fixed prior to the 1970's

After seeing that statement innumerable times on bogleheads and here, it is time to put it to bed.  The only fixed prices were the official government prices.  The U.S. is STILL a fixed price at $42 per ounce, but we only care about the market price.  The market trades at its own price, typically less than the official price if the currency is convertible.  It's complicated.    For reference I've just posted a data source in the "daily gold price" thread of the gold sub-forum with gold prices back a few centuries (London and New York).

[Gumby]

I've seen some posting on the board that look at the PP's volatility on an intra-day basis - which flags to me that those investors are likely to sooner or later be disappointed and might jump out of the PP as a result.

Clive, I have an enormous amount of respect for you. But, I'm not sure that it's more harmful than relying on Simba's spreadsheet to discover and recommend a "better" approach to the PP, based on past market conditions that will likely never happen again. I would think one would be far more likely to jump out of a never-previously-used-before Simba-generated portfolio than the Permanent Portfolio.

As Harry Browne wrote in, Why The Best Laid Investment Plans Usually Go Wrong...

"If anyone had found the magic key to investment riches, he wouldn't be telling you of the profits his system would have produced (hypothetically), he'd be telling you of the profits it did produce."

[Gumby]

Thank you, Clive. That was a very good explanation of where you are coming from — and I think you make some excellent points.

What you've highlighted is simply a reality of investing in general. Harry Browne didn't design the Permanent Portfolio to generate wealth. He was very firm in his belief that wealth-building came from one's career — not one's investments. That being said, a Variable Portfolio (funded with money that one can afford to lose) seems like a much better way to make wealth-building speculative bets.

Harry Browne strongly believed that there needed to be a psychological wall between one's Variable Portfolio and one's Permanent Portfolio — so that one wasn't tempted to change their long term strategy no matter what the future held.

Here's what Harry Browne said of the PP/VP division in Why The Best-Laid Investment Plans Usually Go Wrong:

---
The first step in the search for safety is to separate your investment capital into two different portfolios. I call them the Permanent Portfolio (for investing) and the Variable Portfolio (for speculating).

The Permanent Portfolio is a permanent, balanced collection of long-term investments. Its main purpose is to assure that you're financially safe no matter what the future brings. Once established, it remains virtually unchanged for many years — even as the economy bounces from recession to prosperity to inflation. If properly arranged, it should produce a profit; but its primary goal is safety.

The Variable Portfolio is a separate, changeable portfolio, funded with money you can afford to lose. Its contents vary as conditions change. With the Variable Portfolio, you attempt to beat the market and earn a sizable profit by acting on your judgements about investment trends. You vary the portfolio's investments as you see the opportunities for profit changing.

It is vital that the two portfolios be treated separately when you make investment decisions. The choices you make for one have nothing to do with the decisions you make for the other. Each portfolio has its own purpose and its own rules. They are as unrelated as the portfolios of two different investors.
---


Clive, those Simba-generated PP-hybrids rarely incorporate this critical psychological division into them. Why not devise a cohesive PP/VP strategy that PP investors can comprehend, stick with, and choose how much to risk with?

[moda0306]

Jesus Clive you put a lot of effort into all this. You really bring about the best "but what if" and "but see how" arguments I think one could make against the PP.  I hope all the constant opposition (always friendly & constructive) doesn't discourage you, because I think you provide excellent insights on the markets, history, and even if people may disagree a bit on your PP assessment, they can probably take a lot away regarding what their VP options could be.

PS, Where did you get that bond price calculator?  That's something I've been wanting to have access to for a while.

Thanks again!

[MediumTex]

Clive is one of those people who contributes much more than he probably takes away.

I really appreciate everything he does here.  Aways informative and provocative.

[Gumby]

Clive is one of those people who contributes much more than he probably takes away.

I really appreciate everything he does here.  Aways informative and provocative.

Hear, hear!

[moda0306]

I have to agree with Clive in the sense that it's difficult for me to separate the PP from my VP.  For the while I was toying with weighted PP's (towards stocks, away from cash), or RSPP's.

I don't do either now, but it's hard for me to keep them cognitively separate.

[Gumby]

Thank you so much, Clive. Always informative.

I see the attraction to a single-portfolio approach — and your portfolios always look very tempting — but I think it would concern me if a Japan situation happened where stocks drop for two straight decades, or more.

I guess I wanted to both have my cake and eat it, and originally saw the PP as such an option. 4% real reward with low risk (year down's of less than 10% and relatively quick recoveries when down's are encountered). Due diligence however has led me away from believing that the PP fulfils that role.

Clive, when you saw that the Permanent Portfolio doesn't achieve a 4% real return, is it at all possible that your calculations weren't complete? What about total real return (i.e. with dividends reinvested)? What about rebalancing at 35/15 rebalancing bands instead of rebalancing annually on Dec 1st? I'm sure you've thought about these sorts of things, but I don't see how Simba's spreadsheets (or other tools) can easily accommodate those specific factors.

When I take Harry Browne's PP Chart from here, which includes all dividends reinvested, and I extract the monthly data points using DigitizeIt v1.5, I'm able to extrapolate Harry Browne's published "Total Return" data for the Permanent Portfolio.

Here is that raw data:

 Harry Browne's Permanent Portfolio Total Nominal Return 1970-2003 (CSV)

How does that compare with your own PP calculations for that time period?

[BRESLOW]

Clive, too much information for the average investor. Can not understand all those numbers.

[Gumby]

Ok. Let me see if I get this straight.

Harry Browne has a chart on his website that looks like this:



Under his chart, he wrote:

"Stock results are for an S&P 500 Index mutual fund, and include reinvestment of dividends. Bond results are for a 30-year T-bond, and include interest received. Gold results are for American Eagle 1-ounce coins. Cash results are for Treasury bills, assuming that a 1-year bill was bought at the start of each year."

The chart clearly shows us that a hypothetical PP started with $100 invested on January 1, 1970 had a value of $2,026 on December 31, 2003.

Harry Browne's published hypothetical gain over 34 years equals a Total Nominal Return of 9.25% CAGR.

According to the BLS.gov CPI Inflation Calculator, $2,026 in 2004 would have the same purchasing power as $416.14 in 1970, and a value of $427.22 in 2003. Since the BLS.gov calculator uses average CPI for the current year, we can split the difference of those two values and assume $421.68 as the Total Real Return for Harry Browne's Permanent Portfolio on December 31st, 2003.

Therefore...

Harry Browne's published results, when adjusted for CPI, show a Total Real Return of 4.32% CAGR (less taxes and fees).

Clive, I don't understand how your calculations are more accurate than what actually happened.

[t-bear52]

As a newbie, I'm having some difficulty following Clive's posts this thread. But I get that you are now no longer believing PP provides twice the inflation rate returns like when you wrote the following:

"The PP has historically provided stock like rewards that generally have been around twice the rate of inflation (fell a bit short in the UK since 1972 with a 6.5% inflation rate and 11.1% PP reward).  As such the PP might be allocated 50% and considered as the 2x inflation investment vehicle such that the total fund value is uplifted with inflation over time.  For the other half if you can lock in a safe 8% for the long term then such a combination might serve equally as well or perhaps better than many of the alternatives."

[Storm]

I think you are both right.  Clive is taking out the bond interest payments from his calculations, because he rightly believes that if the past 30 years have been a bond bull market, it is unlikely that the next 30 years will continue the trend.

We could theoretically have a 30 year bond bear market where interest rates slowly grind higher, which would cause the PP to lag behind that magical 4.36% inflation adjusted return.

It is a bit disingenuous to completely remove the bond interest payments, because even if, for example, the 4.625% coupon 30 year bond I bought last week is not very valuable in a future 7% interest rate world, I'm still collecting the payments, however small they may be.

[Gumby]

2.32% real gain + 2% interest rate decline capital gain benefit.

In the other direction, buying a bond for $2 when cash rates were 5% and selling for $1 when cash rates were 10% results in a 50% loss for the investor which assuming a similar 30 year period = -2.3% annualised capital loss bias effect.

What I am suggesting is that over the full interest rate cycle, the average annualised rewards for the PP appear to come out at around 1.6% type amounts ABAICT.

If only it were that simple. The entire 1970s was a bear market for Long Term Treasuries and the Permanent Portfolio still managed to maintain its Total Real 4.32% CAGR the entire time. Harry Browne firmly believed that Gold was powerful enough to offset a very significant loss from Long Term Treasuries. In other words, if Long Term Treasuries drop by 50%, Gold would rise by 200% or 300% or more. I don't see how you could test that theory using your averaging calculations.

Clive, it seems to me that you are simply saying is that you don't like Long Term Treasuries.

[MediumTex]

From a longer term perspective (100+ years), the 1970s bear market in bonds was an anomaly.

Typically, long term interest rates through history have oscillated around a mean of about 4%.

What this means is that interest rates right now are not really low, they're about average.

[clacy]

As long as the US continues to send dollars overseas to purchase goods, those dollars will continue to come back to the US in for form of treasuries and/or US stock market purchases, IMO.  That is from Harry Browne's radio show and it makes sense to me.  That would more or less give us some sort of combination of low Treasury rates and/or high stock prices. 

[Gumby]

Gumby

I don't see how you could test that theory using your averaging calculations.

since then I've gone back a lot further by using silver as a poor man's gold proxy and discovered the 1.65% real average figure.

You can't use Silver in place of Gold. It doesn't work. Harry Browne discussed this on his radio show. He said that Gold reacts much more powerfully than Silver does when interest rates rise, in an inflationary environment. That's why he chose Gold and not Silver. Perhaps that's why you're not seeing what Harry Browne saw in the Permanent Portfolio.

[clacy]

The fact is we none of us will ever really know how the PP will perform in the future.  40 years is not really long enough to draw concrete conclusions.  The HBPP seems to be as solid as it comes in terms of protection, but that protection likely comes at a price in terms of growth.  Which is exactly why I have a significant portion of my investable assets in the PP.  With that said, I have 2/3's in what most here would call my variable portfolio, because I'm hoping to beat 9% pa.

Should I have 100% in my VP or move it all to my PP???? That question won't be answered until after the fact.  Time will tell I guess.