Simba's Spreadsheet from Bogleheads

[Exocet]

Hi all

Please let me start by saying that I'm a PP converted and fully invested in the orthodox variation of the PP per HB recommendation.  However, I constantly keep on reading and analyzing various asset allocations and long-term results.

As most of you are aware, the Simba spreadhseet available from Bogleheads is an excellent resource for scenario building.  What I find it somewhat disturbing is that the 1985-2010 performance of the PP is significantly below that of almost every single other lazy portfolio.  The PP has higher Sharpe and Sortino ratios, because they reflect PP's conservative allocation.  PP has a much better performance than most when we consider the last few years but this makes me wonder if, by investing in the PP, I'm sacrificing long-term results in the pursuit of capital protection in extreme conditions like 2007-2010.

Could anyone please comment on what appears to be quite a low performance of the PP from Simba's spreadhseet, and explain why the PP is a good long-term choice even when results are significant lower.

Many thanks.

[AdamA]

...makes me wonder if, by investing in the PP, I'm sacrificing long-term results in the pursuit of capital protection in extreme conditions like 2007-2010.

I suspect that this is correct. 

Who knows what will happen going forward, though.

[cabronjames]

my off-hand speculation.

1980-1999 were great for stocks.  Many "lazy ports" are heavy stock 50+% of the port, they have "recency bias".  AFAICT OTOH Harry Browne takes the long term 100+ years view of what did happen historically & what might happen based off economic conditions to create the HBPP that beats inflation/maintains purchasing power in most cases, or at least more cases than a "stock bug" lazy port would.

Would be interesting to see that same 1985-2010 window, using those lazy port allocations but with Japanese assets in Japanese Yen terms.  The Perm Port probably would be at or near top of list.  IIRC MediumTex just noted that the Japanese stock market, nominal terms without considering div reinvest, is still below its 1989 high!

[WildAboutHarry]

Simba's spreadsheet uses VUSTX which isn't as strong as 30-year treasury bonds.

That may be part of it.

[MediumTex]

If you look at a lot of the lazy portfolios that compete with the PP, I think many of them were conceived during that amazing secular bull market for stocks from 1982-2000, and thus probaly have more equity exposure than someone interested in safety and stability ought to have in their portfolio.  I think that many of these lazy portfolio users have an incomplete understanding of how much risk they are really taking, as many of them learned in 2008.

If the PP returns from 1982-2000 bother you, you can probaly rest easy, since bull markets for stocks like that are normally once in a generation events at the most (and sometimes they only come along once every few generations).

[Gumby]

Backtesting from way before the closing of the gold window to present day is fairly precarious since the gold standard economy looks nothing like our present-day fiat economy. Our currency is completely different (from gold/quasi-gold standard to complete fiat money). We will almost certainly never have those kinds of currency conditions again.

Here is an updated monthly performance chart of the PP vs the S&P 500 since 1970, adjusted for inflation with CPI data. The data from 1970 to 2003 came directly from Harry Browne's own published data (on HarryBrowne.org), where Browne recorded a nominal 9.25% CAGR. (4.32% CAGR when adjusted for inflation). So, this is probably as accurate as you're going to get in terms of tracking the PP's performance with dividends and proper rebalancing...



Keep in mind that the object of the PP isn't necessarily to beat the market. Rather it's to provide a steady and stable ride that doesn't make you want to vomit and quit your plan at the worst possible time. Most of those other lazy portfolios can be very difficult to stomach during downturns and very few people can stick to the top lazy portfolios in practice when things are going well.

You're not going to strike it rich with the Permanent Portfolio — that's not why we're here.

Despite all of this wonderful data, backtesting doesn't tell us anything about the future.

[Jan Van]

Keep in mind that the object of the PP isn't necessarily to beat the market. Rather it's to provide a steady and stable ride that doesn't make you want to vomit and quit your plan at the worst possible time.

Exactly...

If you like more risk, you can spice things up in your VP.

[Gumby]

Despite all of this wonderful data, backtesting doesn't tell us anything about the future.

You do realise that a large chunk of Harry's chart was based on backtest data Gumby.

Of course. But, the difference is that more than half of Harry Browne's data was the actual recorded performance from his own Permanent Portfolio record keeping. And I suspect Browne's data is a more accurate representation of the PP (versus Simba's spreadsheet).

Showing a backtest from 1926 tells us practically nothing — especially when most of your chart took place during the Bretton Woods System, which will never happen again.

Either way, I think we can all agree that none of these charts tell us anything about the future.

[MediumTex]

I always like Harry Browne's observation that backtesting isn't done to confirm that a strategy works, but rather to determine whether it doesn't work.

If backtesting doesn't invalidate an investing theory, then one has cause for more confidence that the underlying theory has worked in practice the way the model suggests it should.

What is important, though, is to have a good theory before you start looking at backtesting at all.  I think what is more common is to backtest until you find something that has worked for a long time and then attempt to develop a theory that explains why it has worked so well.

[Gumby]

But, the difference is that more than half of Harry Browne's data was the actual recorded performance from his own Permanent Portfolio record keeping.

Which is now simply historic A tad ironic to criticise one historical snapshot as telling us nothing about the future whilst at the same time presenting another historical snapshot. Come on Gumby - show us one for the next 20 years instead

My only point was that Harry Browne's data is more accurate than Simba's Spreadsheet in terms of showing the past performance of the PP. But, I know don't know how many times I have to say it. I don't believe any of these charts show us anything about the future. My criticism about your chart is that it shows the PP mostly during the gold standard era — which makes little sense in our fiat world.

The chart was primarily for a PP holding silver instead of gold. I later overlaid a PP with gold where the gold price was based upon supply/demand price changes (which varied over the Bretton Woods years).

Which proves what exactly? I don't think it would have made sense for anyone to devote 25% of their wealth to fight inflation while the US was on the gold standard. Of course the returns would have lagged. It shows us nothing about the world we live in now — which couldn't be more different.

[Gumby]

Historically it would appear that a stable and relatively safe, albeit lower reward can be achieved by holding some business exposure (stocks), some commodity exposure (gold), some loan exposure (long dated treasury's) and some cash.

You could have just said that from the get go!

[cowboyhat]

One advantage to the stable rise in the PP is that you can get on or get off the ride at pretty much any point. That means you don't have to time your retirement to fit the value of your stock holdings. A huge variable in stock portfolio performance is when you start investing and when you stop. And serious problem if, for example, you planned to retire at the end of 2008.

[Exocet]

Thanks for the responses.  I guess the overall conclusion is that, in bull markets, the PP underperforms but in bear markets it performs better than most asset allocations.  The question is which of the two, either secular bull market (like 1982 to 2000) or secular bear market (2001-2011), is the outlier?  If we accept that both are 'normal' parts of long-term cycles, then the only reason to invest in the PP is risk avoidance and/or  capital preservation.

Interestingly, techniques such as the 200-day MA (as proposed by Mebane Faber) have not worked with PPRFX because of its low volatility - the PP is the only asset allocation technique (other than 100% cash) for which no risk minimization strategy works because risk is already low.

After all, the PP is a clear demonstration of low risk = low reward!

[MediumTex]

After all, the PP is a clear demonstration of low risk = low reward!

This may be more a matter of perception, but to me 9%+ returns are not "low reward".

[rhymenocerous]

The following example really helped solidify for me the importance of stability in portfolio returns.  Highly volatile portfolios can actually increase your risk and lower your returns compared to a more stable portfolio.  It doesn't matter that "expected returns" for stocks are greater than that for bonds.  That doesn't mean that you'll wind up with more money in the end with a 100% stock portfolio as opposed to a mix of stocks and bonds (or in the case of the PP: stocks, LTTs, cash, and gold).

If you have a $10k portfolio that earns 5% in year 1 and 5% in year 2, at the end of the second year you will have $11,025 with a CAGR of 5%.

If you take that same $10k and choose to take on more risk, say earning 20% in the first year, but losing 10% in the second, you'll end up with $10.8k and a CAGR of 3.9%.

As a result of taking on more risk (ie higher standard deviation) there are years where the returns are 15% greater than in first scenario!  However, the increased volatility leads to larger losses in some years and ends up lowering the CAGR overall.

I also like to think of it this way.  Say that same $10k earns 30% in the first year and then loses 15% in the second.  You end up in roughly the same place as the first scenario with $11,500 and a roughly 5% CAGR.  If you end up in the same place, which portfolio would you rather hold?  I'd vote for the stable 5% per year (the PP) over the wild swings in the other scenario (the traditional equity heavy portfolio).

[moda0306]

What amazes me is that even when certain aspects of modern portfolio theory show resounding success (noncorrelated assets (hopefully with macroeconomic fundamentals like in the PP) can be "more than a sum of their parts"), the Warren Buffets and financial planners of the world don't see the value of even SOME LTT bonds or SOME gold in a portfolio.

I don't care how good you are at picking awesome dividend stocks... there'll come a day when the wheels of the macro-economy throw the price way down (Berkshire Hathaway anyone?...), and owning assets fundamentally poised to zig when they zag, in some minimal weighting (5%? 10%?) will stabilize your portfolio, and make it grow more than it otherwise would... EVEN if those assets return less individually.  CAGR and correlation are the factors, not just the former.

rhymenocerous,

Gold and stocks, when equally weighted and rebalanced over the last 39 years, have returned over 11%.  That's the power of proper diversification.  Neither of those two assets returned 11% on their own, and both, individually, had much higher volatility than the combination of the two.

Yet you still get comments by Warren Buffett to the effect that diversification is NOT NECESSARY if you can pick stocks well... Maybe they're LESS necessary (25% weightings being a bit much), but diversification will still improve and smooth your returns, even if it's used in a much more reserved nature.

[Exocet]

After all, the PP is a clear demonstration of low risk = low reward!

This may be more a matter of perception, but to me 9%+ returns are not "low reward".

Simba's spreadsheet shows a CAGR of 8.05% for the period 1985-2010.  Although I'd agree that such a result is not bad at all, it is in fact the lowest out of 25 lazy portfolios.  Even a plain Vanguard Wellesley fund has a CAGR of 10.82% with similar Sharpe ratio.

[MediumTex]

After all, the PP is a clear demonstration of low risk = low reward!

This may be more a matter of perception, but to me 9%+ returns are not "low reward".

Simba's spreadsheet shows a CAGR of 8.05% for the period 1985-2010.  Although I'd agree that such a result is not bad at all, it is in fact the lowest out of 25 lazy portfolios.  Even a plain Vanguard Wellesley fund has a CAGR of 10.82% with similar Sharpe ratio.

If you take it back to the early 1970s, though, the PP provides a much more stable inflation-adjusted return.

The PP is definitely the tortoise in a field of hares.

The ironic thing about many PP tire kickers is that they resist the PP because they perceive it to be too risky, while other more experienced investors resist it because they perceive it to be too conservative.  It's kind of ironic.

I would only say what I have always said on this topic, and it is that the PP is not for everyone, but for those who find that it matches up with their worldview and risk tolerance, it can provide good service.

[doug6zj9]

Simba's spreadsheet shows a CAGR of 8.05% for the period 1985-2010.  Although I'd agree that such a result is not bad at all, it is in fact the lowest out of 25 lazy portfolios.  Even a plain Vanguard Wellesley fund has a CAGR of 10.82% with similar Sharpe ratio.

IF it were 1985.... and IF you could somehow had known this... and IF you had a portfolio that you somehow knew would not be redeemed until 2010.... this would be good to know.