Can the PP perform well when two of its asset classes are falling

[MediumTex]

This is where we come full circle back to the original question of the OP.  If I look at Clive's table I see that it is very clear that the PP cannot perform when two of it's assets are under-performing. 

The question then is, how likely is this for the future?  Nobody can predict what will happen or when, but we can speculate some can't we  

One of the core concepts of the PP is that when stocks fall, LTT rise and vice versa.  If this relationship breaks, what does it mean for the PP?  Is it possible for this correlation to no longer hold?

The PP is simply a theory that has matched up well with reality for the past 40 years.

Will the theory continue to match up with reality going forward?  I don't know, but I do know that as long as the PP continues working I will continue enjoying it.  When the PP stops working (as measured by a rolling three year period of negative average returns), I will certainly reconsider the prudence of following it.

To the question of how the PP will do if two of its assets are falling in value, isn't that what happened in the 1970s as we saw simultaneous secular bear markets for stocks and bonds?  The PP had some of its biggest years in its history during that period.

How will the PP do when three of its assets are falling in value?  We saw that in 1981 and the PP did fine (it had a small loss in a year when almost everything had losses).

The thing is, time is passing as we have these endless PP "what if?" sessions.  As that time passes, you have to do something with your money.  I don't think that it is a sufficient reason not to use the PP today just because you can conceive of some future scenario where it wouldn't perform as well as it has in the past.  Each person has to make up his own mind with respect to this matter, of course.

PP objections tend to come in three flavors:

Flavor 1: "The PP doesn't actually work."  This objection is fairly easy to overcome by demonstrating that it is factually incorrect from a historical perspective.

Flavor 2: "The PP shouldn't work."  To this I think a more nuanced response is needed.  Ultimately, if we are to be realistic and objective, whether something should or shouldn't work is highly subjective and ultimately must yield to what does work.

Flavor 3: "The PP is about to stop working."  To this objection I can't formulate a complete rebuttal because no one knows what the future will hold.  I would suggest, however, that you could say the same thing about any investment strategy and sometimes you would be right and sometimes you would be wrong.  Bailing out of a strategy, however, because someone thinks it may stop working is, IMHO, not the kind of approach to investing that is likely to reap satisfactory long term results.

The PP is not a one size fits all approach to investing.  For those it does fit, however, it can provide good service and a low stress investing experience.  In having these discussions, it sometimes feels like we are holding the PP to an almost absurdly high standard.  I think a lot of these objections are more properly directed to the countless investment strategies that have failed in practice and whose theories have been invalidated through experience, and yet their proponents continue to insist that such approaches are still sound investment strategies.

[AdamA]

The PP is simply a theory that has matched up well with reality for the past 40 years.

Will the theory continue to match up with reality going forward?  I don't know, but I do know that as long as the PP continues working I will continue enjoying it.  When the PP stops working (as measured by a rolling three year period of negative average returns), I will certainly reconsider the prudence of following it.

To the question of how the PP will do if two of its assets are falling in value, isn't that what happened in the 1970s as we saw simultaneous secular bear markets for stocks and bonds?  The PP had some of its biggest years in its history during that period.

How will the PP do when three of its assets are falling in value?  We saw that in 1981 and the PP did fine (it had a small loss in a year when almost everything had losses).

The thing is, time is passing as we have these endless PP "what if?" sessions.  As that time passes, you have to do something with your money.  I don't think that it is a sufficient reason not to use the PP today just because you can conceive of some future scenario where it wouldn't perform as well as it has in the past.  Each person has to make up his own mind with respect to this matter, of course.

PP objections tend to come in three flavors:

Flavor 1: "The PP doesn't actually work."  This objection is fairly easy to overcome by demonstrating that it is factually incorrect from a historical perspective.

Flavor 2: "The PP shouldn't work."  To this I think a more nuanced response is needed.  Ultimately, if we are to be realistic and objective, whether something should or shouldn't work is highly subjective and ultimately must yield to what does work.

Flavor 3: "The PP is about to stop working."  To this objection I can't formulate a complete rebuttal because no one knows what the future will hold.  I would suggest, however, that you could say the same thing about any investment strategy and sometimes you would be right and sometimes you would be wrong.  Bailing out of a strategy, however, because someone thinks it may stop working is, IMHO, not the kind of approach to investing that is likely to reap satisfactory long term results.

The PP is not a one size fits all approach to investing.  For those it does fit, however, it can provide good service and a low stress investing experience.  In having these discussions, it sometimes feels like we are holding the PP to an almost absurdly high standard.  I think a lot of these objections are more properly directed to the countless investment strategies that have failed in practice and whose theories have been invalidated through experience, and yet their proponents continue to insist that such approaches are still sound investment strategies.

Nicely said.  I hope you put it in the book. 

[moda0306]

Clive,

I don't think most people expect the PP to perform the same way over the next 40 years as it did the last 40, but outside of cash (which we know probably won't drive the portfolio), gold and bonds are the ones most commonly assumed as over-priced.

That leaves us with investing in stocks when the earnings yield of the S&P is under 6% if you use 10-years-back earnings.

How about the other options out there:

TIPS: With a negative nominal yield?

Corporate bonds: Callability will neuter anything long-term, and short terms are like lucrative cash.

Emerging Markets: Tempting, but with little respect for private property, political instaibility, and populations with bad distributions of wealth, I don't know if this is anything but a flip of a coin.

Foreign stocks/bonds:  Europe seems to be toast, to me anyway.  I don't see how this could play out well.

REITs:  Could do ok, but are build on leverage... what if rents drop as over-leveraged families start renting basements out or keeping kids home longer?

I just don't see other options giving us much over the next 10 year... can't really predict 40 (though resource scarcity could make things really shaky).

[craigr]

I'd question the comfort factor for someone who started a PP in the early 1980's and saw PP growth not even keep pace with STT's for the next two decades, and who may have been drawing an income (retired). Providing you however have a realistic expectation of the PP being just an inflation pacing investment then fine. There are other safer alternatives to achieve that objective. For instance Harry suggested 100% T-Bills as a second choice option after the PP.

Understand that an investor that simply bought ST Treasuries in 1981 pulled down 8.6% a year and had to take not only no bond risk, but also no stock risk for 20 years.

So by this logic, an investor should just buy ST Treasuries because other portfolios could do worse on risk adjusted basis.

But even then this isn't the relevant comparison because the Permanent Portolio is also focusing on real returns over all these time periods. And even in the worst case you outlined from 1981-2001 it still pulled in real returns approaching 4%.

So even the worst case when picking the peak of peak for gold and buying in all at once the portfolio still did better in real-terms than probably most stock/bond allocations in the 1970s and 2000s where they returned close to zero or below in real returns the entire time.

I'm sure if you had approached and investor in 1981 and told them that for the next 20 years they could get an almost guaranteed 9% return from just owning ST treasuries or could take their chances somewhere else that many would probably go for the sure thing (assuming they didn't think the double digit inflation then was going to continue). Or if they knew that stocks were going to turn in 14% a year over that period they'd have selected that option.

But nobody had that knowledge.

And again, even with that knowledge now after the fact it still didn't matter because the portfolio still performed within spec. That was real returns in the 3-5% range with no significant losses.

And if you simply start your analysis in 1982 the PermPort beats the ST bonds. And if you started it a couple years later it handily beats it.

These comparisons are heavily time dependent and still don't disprove the basic idea which is that the portfolio has had real returns and good protection.

A period like 1981-2001 may come again, who knows? But we won't know until 20 years after the fact and that doesn't mean we are going to pick a better allocation that could take advantage of it.

[MediumTex]

A four factor over 40 years is equivalent to 3.5% annualised. And the benefit of generally transitioning from relatively high to relatively low yields applies to each of stocks, LTT and STT.

But what about rising yields throughout the 1970s?  The PP still did fine.

The classic PP, since 1980 provided a 4.25% annualised real reward. Much of that might generally be put down to financial gravity. Flatten or reverse direction the gravitational effect and the PP runs with a real risk of returning negative real rewards possibly extending over one, two or maybe even more decades.

But that's not what happened when interest rates actually rose over a multi-year period in the 1970s.

I'd question the comfort factor for someone who started a PP in the early 1980's and saw PP growth not even keep pace with STT's for the next two decades, and who may have been drawing an income (retired). Providing you however have a realistic expectation of the PP being just an inflation pacing investment then fine. There are other safer alternatives to achieve that objective. For instance Harry suggested 100% T-Bills as a second choice option after the PP.

In recent years I don't think that T-bills would have been nearly as satisfying for a retiree as the PP would have been.

[systemskeptic]

LTT and cash average out to have a > 5-7 duration depending on the current rates, so in effect 50% of the PP is a leveraged bet on falling rates. The other 50% then becomes a 50-50 bet on prosperity (stocks) and political/economic instability (gold).

Doesn't this leave an absence of any kind of hard counter/bet for rising rates?

[craigr]

Doesn't this leave an absence of any kind of hard counter/bet for rising rates?

Sharply rising rates are good for gold.

Mildly rising rates that stabilize are good for stocks.

[moda0306]

craigr,

Sharply rising rates as a result of spiralling default risk would, I'd agree, be good for gold.

Sharply rising rates on treasury debt as a result of contractionary monetary policy to strengthen the dollar and sap inflation from the economy, like evidenced by 1981, would probably be quite bad for gold, IMO.

You think this is an accurate nuance?

[systemskeptic]

Sharply rising rates are good for gold.

Mildly rising rates that stabilize are good for stocks.

I'm trying to wrap my head around this.  I thought it was rare for gold and stocks to both be rising in the same environment.  Either they raise the rates too fast and stocks crash due to the loss of cheap money, or they raise them just right and we return to an era of prosperity which strengthens the dollar and drops gold.

Either way, these are soft counters that perhaps dependent on a lot of factors, aren't they?

[MediumTex]

LTT and cash average out to have a > 5-7 duration depending on the current rates, so in effect 50% of the PP is a leveraged bet on falling rates. The other 50% then becomes a 50-50 bet on prosperity (stocks) and political/economic instability (gold).

Doesn't this leave an absence of any kind of hard counter/bet for rising rates?

The economy will always be in the midst of some combination of inflation, deflation, recession and prosperity.

There are no other monetary and economic conditions to choose from.

The PP provides you with protection in every combination of these economic and monetary conditions.

If rates are rising because stocks are doing well, your stock allocation will protect you.

If rates are rising because people fear future inflation and/or government default, gold will protect you.

If rates are rising because the Fed wants to slow the economy down (i.e., a "tight money" recession), then cash will protect you.

In each case of potential interest rate increases, the PP protects you. 

[craigr]

craigr,

Sharply rising rates as a result of spiralling default risk would, I'd agree, be good for gold.

Sharply rising rates on treasury debt as a result of contractionary monetary policy to strengthen the dollar and sap inflation from the economy, like evidenced by 1981, would probably be quite bad for gold, IMO.

It is, but the Fed can only raise rates to 1981 levels for so long before causing other problems. The prime rate was over 20%!

So they will either get the problem under control or tip the economy over into a depression! That kind of recession doesn't last long. And after 1981 bonds posted like a +40% gain and stocks also sharply recovered. You just have to wait it out when the Fed does stuff like that.

[craigr]

Sharply rising rates are good for gold.

Mildly rising rates that stabilize are good for stocks.

I'm trying to wrap my head around this.  I thought it was rare for gold and stocks to both be rising in the same environment.  Either they raise the rates too fast and stocks crash due to the loss of cheap money, or they raise them just right and we return to an era of prosperity which strengthens the dollar and drops gold.

Either way, these are soft counters that perhaps dependent on a lot of factors, aren't they?

The Fed messing around with rates can only do a few things: Cause inflation, cause deflation, or put everything in balance. I don't know what will happen (and I'd argue they don't either). I just own an asset that tries to do well in each case. That's all I can do.

If there is an asset that does better in a time where cash is doing best I'd love to know about it. I'd use it in place of cash. But I've (nor Browne) found one that was consistently good in that condition. If you have any ideas I'm all ears.

[D1984]

The OP is gone for a day and a half and the thread gains almost three pages... lots of good points I want to address.

Stone

As regards equality  (and Japan vs the Anglo-American economies in this matter): Maybe it would be better if we were as equal as they were...maybe not. They are a totally different ( rather ethnically/culturally homogeneous and don't make waves culture as well vs what we have in the States) it's probably a moot point...although it's interesting to note that much of the seeds of Japan's equality (few widespread wealth disparities, low CEO-worker pay ratios, workers on company boards or at least directors who had formerly been workers and not hired outsiders, high taxes and greater employment security than the US, etc) were actually imposed to a large extent by the MacArthur occupation government (he was a believer in the New Deal and in Henry George's land tax ideas and was worried about concentrated land ownership, about what would happen if the zaibatsu were allowed to remain as they were before the war and possibly fund hard-nationalist political parties, and about what Japanese workers might do-i.e. support Communism and the Japanese Communist Party- if they felt they weren't getting a fair shake).

As regards swaps: I don't know if they are available to average investors. The vanilla swaps (merely swapping LT rates for ST or vice versa) are available as standard, regulated products that have many willing potential counterparties and in some cases IIRC are exchange traded. Not sure what the minimum is but if enough people got together with several million they could probably form (perhaps in Britain, the Channel islands, or Bermuda if US regulations made this too difficult) their own cash fund/ST fund that was swapped. Of course, if you can find a willing counterparty (and you both have enough treasuries as collateral) you can do an over-the-counter swap for any amount big or small but most swaps are going to be for large amounts. For the record, "turbo swaps" are  "flavored" (i.e. non-vanilla) and "exotic" swaps and such are OTC swaps and are typically customized between counterparties; in other words there isn't (for instance) a standard "pay 30-yr squared, get LIBOR squared" swap rate history like we have (the Fed maintains it on their website) for vanilla fixed-f0r-floating 30 yr to ST swaps.

As regards silver/zeros and wasting my time: I don't consider it a waste of my time at all and in fact it was rather interesting to see how even when correlations were LESS (more negative) from 1952-71 between the two they lost out to inflation if rebalanced or not rebalanced but when rebalanced from 1972-2011 when correlations were MORE (but still negative) they killed inflation and provided decent returns albeit with some hair raising volatility from 1979-1982. Besides...I already had the yearly performance for silver and zeros (synthetic zeros from 1952-1985) so it was only a matter of running an OpenOffice spreadsheet for rebalancing.

As regards gold and its value in non-inflationary times vs its value as Indian jewelry: Who knows....all we know from 40+ years of data is that gold historically does NOT do well when cash provides yields at least a little above inflation and when "good times" prevail economically. Also, while Asians may have been poorer overall in the 50/60s than they are now, there WERE thriving free markets (i.e. didn't have to be priced at $35 an ounce) in the Philippines, Thailand, Macau, and Hong Kong (and Bombay IIRC...and there was also an unofficial one briefly in Taiwan early in this period) so it's not like Asians weren't buying at least some gold back then either.

doodle

As regards 5-year ladder vs TLT....you tried that already...didn't work out so well in falling rates but would in rising rates..the issue is we just don't know WHEN rates will rise.

As regards the 40% 1-5 year ladder and 10% EDV instead of 25/25 STTs and coupon-bearing LTTs idea: I tried a backtest of that (well, sort of...I used 37.5% STTs and 12.5% zeros) from 1952-67 (the years I was concerned about the PP underperforming); it added about 17-18 basis points overall (to the whole PP for 52-67). Not bad and a good idea if you want to "tilt" your Treasuries in the PP instead of a pure barbell but it doesn't come close to turning a 2% overall real return for the whole PP into the 4-5% real returns that it has delivered from 1972-2011.

Concerning CDs vs LTTs - Again, not a bad idea by itself but it does mean giving up some of the deflation protection of LTTs (then again, if LTTs fall to 1% yields or so I am moving a small portion of my own LTT exposure to shorter terms as the potential loss in LTTs is much greater than the potential gain at that point IMO).

I also want to take this as an opportunity to ask you to consider (and this is also to everyone else-MediumTex, Craigr, MachineGhost, Stone , moda, Systemskeptic, etc...who have mentioned in this thread LTTs and the effect rising rates have on them) that the issue per se is not that "LTTs will be harmed when rates rise." We know that and expect that; they aren't here to do well when rates rise but to do well when they fall, whether that fall is due to deflation a la Japan or to worries about markets in turmoil like in 2008 and 2011. The elephant in the room IMO is that the asset (cash) that is SUPPOSED to do OK in a rising rate environment simply does not have the volatilty that LTTs have when rates fall. No amount of tinkering with how many LTTs or what kind of LTTs you hold will change that. The PP thrives on volatility and IMO what you need is more powerful (more volatile) cash, not less powerful (less volatile) LTTs.


MachineGhost (and doodle)

Concerning "turbo swaps" and other swaps that are highly volatile vis-a-vis interest rate increases/decreases: Yes, they do add some complexity to a cash position of a perm portfolio and maybe aren't in the "true spirit" of the PP but how else can cash be made as volatile on the upside (rising rates) as LTTs are when rates fall? If you (or anyone else) have another idea that can make cash gain 15-20% a year in a rising rate environment like LTTs can when rates fall I'd love to hear it.

As far as paying 12.25% and getting 2.25% is concerned: Three things to keep in mind.

One, that only applies as long as rates stay low. When and if they rise to even 3.5% you break even (in this example) and if they rise past that it's all gravy. If they stay low (or fall further although there is a floor of sorts as if the ST rates falls below 0% and it is squared then for instance -0.1% times -0.1% is still close to zero percent...and most swap agreements would have a 0% floor cap anyway) then you lose 9.75% per year but fixed-for-floating swapped cash is not designed to do well when rates fall or stay low any more than LTTs or zeros are designed to do well when they rise. It is designed to give cash/STTs the same or similar power in rising rates as LTTs have when rates fall.

Two, you don't have to swap all your cash/STTs. Turbo or leveraged swaps can be used in a "tilt" fashion to achieve similar effects on a smaller notional amount as vanilla swaps on a larger amount. Consider the following two choices: One, pay 30-yr rate but receive 3-month LIBOR rate. 30 yr is currently at around 3% and LIBOR is at 0.58%. You could do a "vanilla" swap (a simple fixed for floating one) on 100% of a certain notional value (say $100K) and start to see positive results when and if LIBOR rises to and above 3%. You'd lose about 2.4% each year ($2420...2.5% of the notional amount) if LIBOR and LT rates didn't change If LIBOR rose to 3.5% you'd still be paying 3% but collect 3.5% for a 0.5% gain (actually you wouldn't be paying anything; swaps are "netted" each month or annually so your counterparty would just pay you the $500). If LIBOR rose to 4.5% you'd gain $1500 on your $100K (plus you'd keep any interest the $100K paid). You wouldn't really start to see sharp gains, though, until and unless rates rose to Volcker levels.

Alternatively, you could choose to do a "turbo swap" (squared swap) on only 20% of the same notional value of $100K or $20,000 instead of a vanilla swap on the whole amount. This would net you a loss of around 8.65% (3 x 3 = 9 so you pay 9%; 0.58 x 0.58 = 0.34 so you receive 0.34%; netted this equals you paying your counterparty about 8.65%) but that loss of 8.65% would only be on the $20K notional amount, not all $100K of your cash...so it would only be a loss of 1.73% on your $100K of cash/STTs overall...this is less than the 2.4% you'd lose above. On the other hand, if rates ROSE to 3.0% you'd still break even as above (you pay 9% but get 9% ) but if they rose to say 4.5% you'd get around 20% and still pay only 9% (so you get an 11%gain on $20K...so 2.2% on $100K)...and if they rose to 5% or 6%  ;D

To be fair, the turbo swap might perhaps cost you a little more (say you pay 16% instead of 9%) to compensate for the risk the counterparty is taking but if that was the case then you could just swap on a little lesser amount and if rates rose significantly you'd still do very well. Also consider-if you are thinking that this sounds like a really risky deal for your counterparty and you are asking why would anyone take it-that if rates stay low for years (a la Japan mid 90s to today or the US from the mid 30s to the early 50s) then the counterparty makes a no-sweat 9 or 12 or 17% on STTs (plus whatever interest he earns) and that even if LIBOR rises to 4% (in the 18% example above) he still breaks even. Your counterparty is basically someone looking to "vacuum up nickels in front of a steamroller" and he is the one taking "fat tail" (i.e. low probability of an outcome but catastrophic consequences if it happens) risks while you are basically paying an "insurance premium" in the form of your leg of the swap (the fixed portion) and you will "collect on the policy" or "the policy will pay off" if rates rise significantly.

Three, paying 12% (or 16%, or whatever percent) to get a lower percent (UNLESS rates rise) is speculative BY ITSELF. But then again, isn't owning LTTs or LT zeros when rates are below 3% (you're speculating that rates won't rise...if they do you could lose big like you did in 1980-81 or 2009) speculative? Isn't holding gold when it has risen in value nearly sevenfold and may be in a huge bubble speculative (you're speculating that gold prices will continue to rise and real rates will continue to be negative...if not you could lose 20-30% or more a year)? Isn't buying stocks when their PE10 is still in double digits and the Euro could collapse (and we could see another 2008-or worse-all over again as the worldwide financial system goes to hell in a handbasket) speculative? I submit to you that holding any ONE of these is indeed very speculative but together in the PP they make a very well hedged combination where gains in certain assets more than offset losses in others.

Look...to be honest with you I'd actually prefer a simple yes/no spreadbet system where you can make a bet (on a certain notional amount) and receive maybe 8 or 9% of that notional amount if rates rise and lose the exact same amount if rates fall (regardless of how much they rise or fall). AFAIK no one has come up with such a system (and it may violate some fundamental law of economics besides) so if you want cash to have the punch in rising rates that LTTs have in falling ones then swaps are what we are stuck with unless you can think of anything better to make cash truly perform at double-digit gains when rates rise (especially when rising from a low base like today's).

craigr

The issue is not the Dow (or gold, or LTTs) falling 90% in one day...the PP can probably deal with that and in fact if rates immediately rose to say 10% and killed LTTs then everything would still be OK as long as rates eventually fell (and we got the gains from rising LTTs as they did so). In fact, based on some quick calculations I did we might be better off if the LTT bull ended with such a bang instead of the whimper of slowly rising rates (again, as long as rates fell again eventually or at least if inflation lagged those high rates significantly). The issue is a "death of 1,000 cuts" scenario where two assets are falling and one (probably cash) is barely keeping up (or earning maybe 1% real) with inflation. The one remaining "winner" asset that is doing well (say stocks in the US from 1950-1967, or LTTs in Japan in the 1990s) simply lacks enough punch by itself to make up for two losers (the losing assets) and one slacker (the asset that just paces or barely outdoes inflation).

On 2009 (and this is directed to anyone)

The PP did indeed do well in 2009 despite the fact that rates rose and LTTs got killed. I wouldn't extrapolate this result to any longer period of time for several reasons. One, both gold AND stocks went up in 2009 (stocks up more than 26%, gold up about 24%); if we had rising rates in a non-inflationary prosperity scenario stocks might still do well but gold would probably fall along with LTTs. If you have a year with 26% stock returns, -22% LTT returns, 0% cash returns, but MINUS 24% (or -21% like in 1997...or minus 10% + like we had several times from 1983 to 1991) then basically you have 2009 but with gold doing the opposite of what it did that year. The end result wouldn't be pretty....minus 4% or so before inflation. Losing 4% (or maybe 5-8% after inflation) for one year doesn't concern me (the PP has had some years when it was down 10% or more after inflation); my worry is what happens if LTT yields keep rising and gold keeps falling while cash is barely yielding 1% beyond inflation. If LTT yields rise sharply and fairly quickly (say over 3-4 years) to maybe 5.5 or 6.5% (the latter being roughly where they were as few years ago as 2000) and then reverse the PP as a whole might do OK as falling yields again mean that LTTs do well once more (and combined with stocks in a non-inflationary prosperity offset rapidly falling gold) but if yields just grind ever higher in a slow, inexorable, and (usually) one-way pattern like we saw from 1950-67 then the PP could have a period of two decades (or more) when it just barely beats inflation. That's what troubles me.

I would also like to add that 2009 was also rather unique in that (besides two asset classes rising very strongly) stocks returned over a 26% real (inflation adjusted return) annualized return that year. That can easily happen over one year or maybe two, but to expect something like that to carry the portfolio over a decade or more (if gold and LTTs both fall) is presumptuous and foolish IMO. Neither the 1920s, 1950s-60s, nor 1982-99 bull market saw stocks return anywhere near 26% per year real on an annualized basis.

On the PP doing well in the 70s and early 80s despite rising rate

Craig, didn't you say that 1968-1974 could maybe be disregarded because it perhaps took that long for gold to adjust to free market non-controlled pricing? If we start at January 1975 and end in December 1980 gold returned 21% on an annualized basis (from January 1975 to December 1979 it returned a little over 23% annually). Heck, gold returned 135% in ONE YEAR in 1979. In non-inflationary prosperity stocks will be what carries the PP, and a broad-based index of stocks has NEVER returned anywhere near that much in one year (or AFAIK even in two really good years end-to-end...at least starting in January instead of starting from a market bottom...and 1979 was NOT starting from a market bottom for gold). One also has to give considerable weight to the fact that stocks actually did OK from 1975-1979 (or 1975-1980); they returned about 6.20% real and 8.14% real respectively which are both as high (or higher) than the historical average of a real 6% or so annually. Finally, LT yields actually fell in 1976 so there was at least one fairly good year for bonds during this otherwise awful period for LTTs.

Typically each year the PP will have one asset class doing wonderfully, one doing OK, one just barely keeping up with inflation or maybe losing 1-2% a year after it, and one asset sucking out loud. From 1975-1980 the asset that was (overall) doing very well was gold, the "OK" asset was stocks, cash just barely kept up with inflation (or lost a few percent after inflation) until Volcker started working his magic, and LT bonds got cut to ribbons in real terms. If we do have non-inflationary prosperity at some point in the near future then stocks should do very well (maybe 16-20% a year) but which asset is going to be the one that from current low rates gives 6-8% real yields (like stocks did in the late 1970s) to help stocks compensate for falling LTTs and falling gold? Inquiring minds want to know.

So the conclusion on the 1970s runs something like this: They don't prove that the PP can do well with two assets falling because during the early 70s (from 1970-1974) gold was still adjusting to free market pricing (I'll willingly give you that bonds and stocks did poorly this whole period because while from 1970-1972 both did well in real terms they gave it back and then some in 1973-1974) and thus doesn't really count; you had two assets doing poorly (stocks and bonds) and one that was starting from an artificially low "starting point" if you will. We can only go off the gold standard once so gold will (almost certainly) never be that low again in real terms unless Ron Paul or another hard-money type gets elected (very unlikely). The late 70s (1975-1979 or even to 1980) aren't a good example either because you had one asset doing great (gold), one doing at least as good as it did historically (stocks), one losing a percent or so a year after inflation (cash/STTs) and only one really and truly sucking (LT Treasury bonds). Cash/STTs wasn't IMO enough of a "losing asset" (they lost about one percent a year during this time frame...compare that with stocks from 1968-1974, bonds from 1973-1981, gold from 1988-1997, or even cash itself from 1941-1948 and you'll see what I mean by "enough of a losing asset" ) to be counted as two assets (when combined with LTTs) going down during this period...do you really think LTTs in a rising rate environment will only lose 1% a year or that gold in non-inflationary prosperity with rising real rates will lose that little? Perhaps they will-stranger things have happened and no one can predict the future-but I wouldn't bet my portfolio on it.

On rising rates being a "headwind" (or a "financial gravity" ) for the PP vs falling rates being a "tailwind"

Clive's thesis as regards interest rates seems very persuasive and the simulated backtests I have done for the years 1950-1967 corroborate the theory that rising rates WILL take perhaps 1.5% to 2% off of the 4-5% real returns we expect from the PP. If anyone can refute this (after taking into consideration what I posted about the 70s and considering that long term rates can more easily fall from 15% to 3% than they can fall another 12% to -9%) please do so, but if not then maybe it's time the PP got some more firepower in its cash portion or else time to switch to Clive's SCV tilt/PP blend.

[systemskeptic]

The elephant in the room IMO is that the asset (cash) that is SUPPOSED to do OK in a rising rate environment simply does not have the volatilty that LTTs have when rates fall. No amount of tinkering with how many LTTs or what kind of LTTs you hold will change that. The PP thrives on volatility and IMO what you need is more powerful (more volatile) cash, not less powerful (less volatile) LTTs.

Overall, D1984, I think you make some really good points and it appears you share the same concerns that I do.  One thing to mention about the volatility of cash is: if a vehicle did exist that had the same volatility as LTT when rates rise, wouldn't the two almost exactly cancel each other out while returning the current rates? Two such vehicles each held at 25% would then be the same thing as holding 50% cash.

I think what we must be forced to conclude is that the PP, as it stands, has an intrinsic bias towards falling rates.

[systemskeptic]

I just took the liberty of backtesting a "rates agnostic" version of the PP which uses 50% SHY instead of 25% TLT and 25% CASH and the results were this:

From 11/19/2004 to present.
50/25/25:       8.3% CAGR, 15.5% max drawdown
25/25/25/25:  9.8% CAGR, 15.1% max drawdown

On 11/19/2004 LTT rates were 4.89% so that is just about a 2% drop until present.

[MediumTex]

The elephant in the room IMO is that the asset (cash) that is SUPPOSED to do OK in a rising rate environment simply does not have the volatilty that LTTs have when rates fall. No amount of tinkering with how many LTTs or what kind of LTTs you hold will change that. The PP thrives on volatility and IMO what you need is more powerful (more volatile) cash, not less powerful (less volatile) LTTs.

Overall, D1984, I think you make some really good points and it appears you share the same concerns that I do.  One thing to mention about the volatility of cash is: if a vehicle did exist that had the same volatility as LTT when rates rise, wouldn't the two almost exactly cancel each other out while returning the current rates? Two such vehicles each held at 25% would then be the same thing as holding 50% cash.

I think what we must be forced to conclude is that the PP, as it stands, has an intrinsic bias towards falling rates.

Even though it was conceived during a period of rising rates and had its very best year (1979: 42%) in a rising interest rate environment?

Your theory seems out of synch with what has happened when interest rates were actually rising.

[craigr]

The elephant in the room IMO is that the asset (cash) that is SUPPOSED to do OK in a rising rate environment simply does not have the volatilty that LTTs have when rates fall. No amount of tinkering with how many LTTs or what kind of LTTs you hold will change that. The PP thrives on volatility and IMO what you need is more powerful (more volatile) cash, not less powerful (less volatile) LTTs.

Overall, D1984, I think you make some really good points and it appears you share the same concerns that I do.  One thing to mention about the volatility of cash is: if a vehicle did exist that had the same volatility as LTT when rates rise, wouldn't the two almost exactly cancel each other out while returning the current rates? Two such vehicles each held at 25% would then be the same thing as holding 50% cash.

I think what we must be forced to conclude is that the PP, as it stands, has an intrinsic bias towards falling rates.

Even though it was conceived during a period of rising rates and had its very best year (1979: 42%) in a rising interest rate environment?

Your theory seems out of synch with what has happened when interest rates were actually rising.

PLUS, it was re-swizzeled in the late 1980s after rates had fallen from the very high low-teens to under 8%. Only to have rates go up to almost 9% the next few years.

The portfolio was not biased towards falling rates. If anything, the original portfolio biased towards *rising* rates and inflation. It overweighted gold, silver, real assets and Swiss Francs. None of which are good for falling interest rate environments. The later 4x25 split got rid of that bias.

[D1984]

Overall, D1984, I think you make some really good points

Thank you, systemskeptic

and it appears you share the same concerns that I do.

Yep....rising rates and what they'd do to the PP's returns (and how cash may not be powerful enough to prevent it)

One thing to mention about the volatility of cash is: if a vehicle did exist that had the same volatility as LTT when rates rise, wouldn't the two almost exactly cancel each other out while returning the current rates? Two such vehicles each held at 25% would then be the same thing as holding 50% cash.

It wouldn't do exactly that...if you wanted a vehicle that did EXACTLY what LTTs did but in reverse then you could just use a futures contract or buy a short LTT fund (or 20+ year bear fund). What I mean by "as much volatility as LTTs" is the potential for 10, 15 ,20 or even 30% gains when rates rise. Even if cash/STTs only rose 8-9% per annum (plus whatever it yielded) when rates rose that would be fine. As it stands all; cash has going for it is a rising yield when rates rise (and with a 1-5 yr ladder like FSBIX/FSBAX or a 2-5 year ladder like FIGTX it doesn't even have much of that at current rates as rising yields may be neutered by capital losses at the longer end of the ladder

Another reason they wouldn't "cancel each other out" is that both are expected (or would be) to have a positive return in the long run when rebalanced as assets with opposing volatility and because each would typically decline less than it went up over the run (i.e. the downs in one would overall be lesser than the ups in the other were high). In fact, even if both assets were losing money overall due to "volatility rot" they could produce a positive return when rebalanced together. As a hypothetical (and unlikely) example but one that nonetheless serves to drive the point home, imagine two assets (needn't be cash or LTTs or gold or stocks or anything in the PP...just call them "A" and "B." ) Have each decline 20% one year then rise 24% the next and have the other do the exact opposite. They switch positions (rising 24% or falling 20%) each year. See what results you get after, say, 10 years when not rebalancing vs when rebalancing.  Un-rebalanced the combined value of both assets will add up to slightly less than your starting value (i.e. they both lost money over the ten years) but rebalanced annually you should have an almost 22% cumulative gain. For a real shocker try (if you haven't already) the above with -99% and 100% as the two values...the unrebalanced assets have all vanished (100% loss) in ten years but the rebalanced assets have a slight cumulative gain (a bit over 5%)

[D1984]

The elephant in the room IMO is that the asset (cash) that is SUPPOSED to do OK in a rising rate environment simply does not have the volatilty that LTTs have when rates fall. No amount of tinkering with how many LTTs or what kind of LTTs you hold will change that. The PP thrives on volatility and IMO what you need is more powerful (more volatile) cash, not less powerful (less volatile) LTTs.

Overall, D1984, I think you make some really good points and it appears you share the same concerns that I do.  One thing to mention about the volatility of cash is: if a vehicle did exist that had the same volatility as LTT when rates rise, wouldn't the two almost exactly cancel each other out while returning the current rates? Two such vehicles each held at 25% would then be the same thing as holding 50% cash.

I think what we must be forced to conclude is that the PP, as it stands, has an intrinsic bias towards falling rates.

Even though it was conceived during a period of rising rates and had its very best year (1979: 42%) in a rising interest rate environment?

Your theory seems out of synch with what has happened when interest rates were actually rising.

MediumTex, 1979 was a year when gold rose 136%. If you expect gold to rise that much again then be my guest (that would put it at around $3,900...far above its previous 1980-81 high in inflation-adjusted terms) but I wouldn't expect that again in any environment short of hyperinflation or economic chaos...or perhaps short of some corporation, person, or central bank trying to corner gold the way Hunt did with silver (for that matter consider how much of gold's rise in 1979 was sympathetic with the rise of silver and the attempted silver corner and how much was due to inflation, negative real rates, oil shocks, the hostage crisis, etc).

In non-inflationary prosperity, do you honestly think that the asset class (equities) that is supposed to be the "star" under that set of economic conditions will honestly give that kind of gains in one year so that we could enjoy 42% gains when bonds and gold are both falling?

[MediumTex]

First, before we get too far away from it, that's a great post above.  Very thoughtful and detailed.  Thanks for taking the time to organize your thoughts in that way.

MediumTex, 1979 was a year when gold rose 136%. If you expect gold to rise that much again then be my guest (that would put it at around $3,900...far above its previous 1980-81 high in inflation-adjusted terms) but I wouldn't expect that again in any environment short of hyperinflation or economic chaos...or perhaps short of some corporation, person, or central bank trying to corner gold the way Hunt did with silver (for that matter consider how much of gold's rise in 1979 was sympathetic with the rise of silver and the attempted silver corner and how much was due to inflation, negative real rates, oil shocks, the hostage crisis, etc).

If I had walked up to you in 2000 and told you that the price of gold (then $250 per ounce) would be nearing $2,000 an ounce a little over a decade later, and most of the gains in the price of gold would have occurred against a backdrop of tame or non-existent inflation, what would you have said to me?

If I further told you that in the same year that gold prices neared the $2,000 mark in 2011 that the yield on long term treasuries would dip below 3%, what would you have said to me?

Do these things sound like they would have made any sense at all back in 2000?  They wouldn't have made any sense to me.  Back in 2000, people who talked about gold were considered a little goofy, and even back then the conventional wisdom was that interest rates had gone just about as low as they could possibly go.

But look what actually happened.

In non-inflationary prosperity, do you honestly think that the asset class (equities) that is supposed to be the "star" under that set of economic conditions will honestly give that kind of gains in one year so that we could enjoy 42% gains when bonds and gold are both falling?

It doesn't really matter what equities do.  What matters is what the overall portfolio does.  During periods of amazing prosperity such as 1982-2000, the PP did fine.

I don't know what to say to some of these theoretical objections other than when the situation has actually arisen in practice the PP has taken it all in stride.

It's also important to understand how what actually happens in the future is probably too wacky to try to even ballpark right now.  If you had told people in the late 1970s that inflation was about to disappear for a generation and gold would be a terrible asset to own for the next 20 years, they would have said you were crazy.  If you had asked someone in 1990 after the fall of the Soviet Union whether the U.S. would spend over half of the next 20 years fighting two wars in two different countries simultaneously, they might have said "Two wars? Over ten years?  Like two Vietnams? At the same time?"

Weird stuff happens.  This aspect of the protection offered by the PP should not be underestimated.  As I wrote on the BH board a while back, it's sort of amusing how someone like Nassim Taleb writes a book with the central insight of "lightning strikes in unexpected places at unexpected times", while Harry Browne would say "Well, yes, of course lightning strikes in that way.  It's lightning.  What I would like to talk to you about is a cost effective lightning rod that anyone can use...."

[MachineGhost]

As regards the 40% 1-5 year ladder and 10% EDV instead of 25/25 STTs and coupon-bearing LTTs idea: I tried a backtest of that (well, sort of...I used 37.5% STTs and 12.5% zeros) from 1952-67 (the years I was concerned about the PP underperforming); it added about 17-18 basis points overall (to the whole PP for 52-67). Not bad and a good idea if you want to "tilt" your Treasuries in the PP instead of a pure barbell but it doesn't come close to turning a 2% overall real return for the whole PP into the 4-5% real returns that it has delivered from 1972-2011.

Since 1928, the PP has returned 3.59% real CAGR.  Willing the real return be higher when gold came off the peg within the last 40 years and LT bonds reverted back to the long-term historical average over the past 30 years is probably a one off occurence.

the PP but how else can cash be made as volatile on the upside (rising rates) as LTTs are when rates fall? If you (or anyone else) have another idea that can make cash gain 15-20% a year in a rising rate environment like LTTs can when rates fall I'd love to hear it.

Instead of outsized/leveraged risky cash gains in a rising rate environment on a base of 25% cash, use reallocation to increase the cash base while earning normal safe gains concurrently with less loss drag from the relatively smaller other three assets.  There will be a sweet spot where the $ gain of both scenarios match.  Quick $100K example: 25K @ 15% = $3750 - $15K (20% loss drag) == 62.5K @ 6% = $3750 - $7.5K (20% loss drag).  Okay that doesn't exactly balance out because of the drag, but it should give the general idea.

ribbons in real terms. If we do have non-inflationary prosperity at some point in the near future then stocks should do very well (maybe 16-20% a year) but which asset is going to be the one that from current low rates gives 6-8% real yields (like stocks did in the late 1970s) to help stocks compensate for falling LTTs and falling gold? Inquiring minds want to know.

Unlike the 1970's, stocks are not undervalued currently, so it would be foolish to expect near returns like that going forward unless we first have a gold bubble and bond bear that precipitates another Volcker-style deflationary event.  The Fed cannot raise short term rates without provoking unproductive inflation because it is at such a low nominal rate on such a huge monetary base.  Any move in short term rates upwards will skyrocket the velocity of money as everyone plays hot potato.  It doesn't seem like we're anywhere near that scenario yet, but it is a huge concern to have three bubblelike assets concurrently with cash to pick up the loss drag if they all pop simultaneously.  It's all a question of timing.  I, for one, don't believe the PP has been yet battle tested under such a scenario.  1981 would have been disastrous if stocks were also overvalued rather than despised.

My personal solution, also despised, is to rely on tactical allocation and reduce exposure during regimes where things are, indeed, in a bubble.  But, it has costs of tracking error to the regular PP.  No one can say whether it will be superior or worse because there simply is not enough history available to backtest it.  But, I can sleep better at night not worrying about a 50% real drawdown in a 1981-style triple bubble event or 20-years of real underperformance.

MG

[Roy]

Hi, Clive.  Good to see the monkey.

A few points:  In more traditional portfolios, I understand a position in foreign equities.  But I wonder if the 25% position in Gold means that the PP doesn't need any foreign equities.  It clearly does not "need" any to perform great, and that's just one less asset group to consider.  What do you think?

I agree those two portfolio types share common features and the blend would theoretically improve things further (judged by past performance).  But we can always find some admixture (some quite odd) to be better than the simple thing that has been, and remains, damned good. 

And I wonder if it isn't easier to get behind, and stay behind, a simpler mix with fewer rebalancing issues.  Not sure but I err to simpler as being easier to hold for DYI types (though I agree with Larry that the simple thing can get overstated).  I do think this is one (psychological) advantage that PRPFX may enjoy—its only advantage, in my view. 

[stone]

If I'm brutally honest, I hold the PP as a speculative play on real rates remaining (close to) negative for the future. My household savings are anyway mostly in cash because that is all my better half trusts so for me the PP is a way to counter the erosion of that cash.
I know that the future is unknowable but don't you need some mental acrobatics to imagine how rates are going to be raised when the stock of bank reserves (competing all yields down) is as huge as it is now and still growing?

Steadily rising real (inflation adjusted) interest rates with a government debt to median wage ratio like we have, really is a much bigger deal IMO than people on here are making out.

[moda0306]

stone,

The economics that would force fairly positive interest rates on short-term debt are either Austrian monetary policy getting enacted, or some kind of economic recovery that creates demand for loanable funds, boosting stocks significantly.... the latter seems the only likely scenario, and the former we should be able to see coming (say, if Ron Paul were running for president and looking to have a decent chance).

So while gold and LTT's might suffer, cash will do alright and stocks will do phenominally, at least in the non-Austrian scenario.

At least that's how I see it happening.

[moda0306]

PS,

I'm hedging the LT treasury portion of my portfolio for rising rates by buying EE bonds to hold until their original maturity of 20 years, at which point they double, and hand you a 3.53% CAGR.  This is juxtaposed against a 20-year treasury bond yielding 2.57% today.  This is pretty easy for accumulators if you only have EE's make up a portion of your balance.  Keep in mind, though, it's a bit of a laddering technique, as 20-year treasuries are where you usually sell because they're not potent enough, so I'd consider them partially as cash and partially as bonds.  It eliminates marketability until the 20-year mark, but it also eliminates principle risk, and accrues tax free. 

I'd suggest if that's a concern for any of you that you do the same, as long as you understand the risks.