20% annual returns over 40 years...interested?

[stone]

Machine Ghost, I'm not sure what you're getting at.

[Storm]

Machine Ghost, I'm not sure what you're getting at.

He's just pointing out that it's not some buy and hold passive fund, it's an actively traded fund, more comparable to a hedge fund than a Vanguard type fund.

[Wonk]

It's been exactly 2 years to the day on this experiment.  Here are the updated numbers:

1xPP: 26.47% (0 rebalance)

2xPP: 62.80% (2 rebalance)

Maybe we can have another 2008 liquidity crisis sometime soon so we can see what happens to these funds.  I've also been tracking an ultimate buy and hold (4-fund) PP, a small cap PP, a 2x small cap PP, a 30-30-30-10(cash) PP, and a 3xPP I added to the mix starting in October '11 out of curiosity.  The main observation is the 1x vs 2x PP.  Will report back in January 2013.

[beafet]

How about:

BGU 33%
TMF 33%
UGLD 33%

Checked it today (a little over 2 months from when I started the simulated portfolio at SmartMoney.com), it is currently at 17.60% It fluctuates a percent or two or three most days, but like the standard PP, seems to average an upward trend. Fun to watch anyway.

[modeljc]

Wonk,

What re balance bands would you use for 3XPP, 33% each just for fun.  Is it 46/20?  Anybody out there willing to risk real money? 

[clacy]

I might be tempted to look at a 2x HBPP after a 10-15% draw down in the standard version.

[modeljc]

The follwoing link warns of Leverage Trap for 2X and 3X products .  Several other sites also say that the volativity is what kills the returns over longer periods.  http://kidgas.hubpages.com/hub/BewareLe ... oxicAssets

[Wonk]

Wonk,

What re balance bands would you use for 3XPP, 33% each just for fun.  Is it 46/20?  Anybody out there willing to risk real money? 

The 3xPP would be 25%x4, so the same 15/35 rebalancing bands would apply.  For a 33x3PP(no cash) I don't know...

As for the decay issue in leveraged funds, it may exist.  I haven't seen it over 25 months of tracking the leveraged PP, but it may exist.  As of today the numbers were approx 28% 1xPP and 68% 2xPP.

[modeljc]

Wonk,

I went to Yahoo Historical prices.  I looked at the period Oct. 17, 2011 to Feb. 15, 2012.  GLD was 162.62 and went to 168.11 for a 3.3% increase.  UGL was 92.24 and went to 95.81 for a 3.9% increase.  UGLD was 50.10 and went to 51.38 for a 2.6% increase.  The 3/ product UGLD should be up 9.9%.  I then went to ETF replay and they had GLD up 3.4% and UGL up 3.9%. 

[Wonk]

Wonk,

I went to Yahoo Historical prices.  I looked at the period Oct. 17, 2011 to Feb. 15, 2012.  GLD was 162.62 and went to 168.11 for a 3.3% increase.  UGL was 92.24 and went to 95.81 for a 3.9% increase.  UGLD was 50.10 and went to 51.38 for a 2.6% increase.  The 3/ product UGLD should be up 9.9%.  I then went to ETF replay and they had GLD up 3.4% and UGL up 3.9%. 

Leverage funds are created to reflect 2x or 3x daily movements and not any other way.  On an expanded time horizon with little trending movement, it's easy to see tracking error in isolation.  The question for 2x or 3x PP is what happens to the overall portfolio collectively?  I started tracking a 3xPP beginning Oct 24, 2011 and compared it to the same starting date for a 1x and 2x.  The results as of today:

1x: 4.03
2x: 7.25
3x: 10.13

When the 1x portfolio is collectively trending up, the 2xPP experiences more than 2x performance over an extended timeline.  When the 1x portfolio trends sideways, you see the 2x and 3x fail to perform exactly 2x and 3x of the benchmark.  I guess you could label this decay, but it's really just the result of daily compounding.  We haven't seen these funds in a market panic or tight-money recession where the 1x does not perform well, so there's no way to tell how they'd perform.  I suspect if the 1x returns -4% as it did in 1981, the 2x would return something like -10% as the daily compounding would have the same effect(inversely) as what we've seen when it is trending up.

[modeljc]

Wonk,

On a collective portfolio I got the same numbers that you did: 1X=4.03, 2X=7.3 & 3X=10.1. WOW!  I probably will never understand why this tracks.

[6 Iron]

This post and the one Clive made above it make sense to me. In a low interest rate environment, where is the downside?  Are there other factors in the decay besides borrowing costs for leverage?

With cost of borrowing so low, a 3x PP might achieve 3 times 1x PP rewards.

Scaling for similar potential rewards to a 1x entails holding perhaps something like

8.3% BGU (3x stocks)
8.3% TMF (3x 30 year T)
8.3% UGLD (3x gold)
75% cash

If those that gain rise >3x and those that lose decline <3x !!!

You'll also hit more frequent rebalance points and potentially capture better 'cost-averaging' benefits. Yet you're holding 75% in cash (less overall risk) !!!

[modeljc]

Clive, I think you may on your way to building a super brick Sh*t house in Arkansas.  I would love to have 75% in cash and only have 8.3% in BGU and 8.3% in TMF and 8.3 in UGLD if it would do one time PP.

I was going to post this on the thread that Medium Tex started by asking will there be another tight money recession.  Instead, what if you could be 75% safe and maybe you get something out of the BGU, TMF, & UDLD.  My post to Medium Tex is:

Maybe I can see the bottom of the market and cycle.  I can’t see tomorrow or next year.  The PP will do well and make money until the cycle ends.  I lived through the end of the 1980’s cycle.  I was 32 and had 100% allocation in gold.  I was lucky and did not know the risk I was taking.  I did not make any money, but I did keep what I started with. 

For fun, and if I can see the way this ends, let’s play a rise in interest rates.  Inflation last year was +3.6%.  If you add a return for the long Treasury bond you get maybe +2%.  I don’t know when or what will cause this rise.  Let’s play the TLT yields 5.5% to 6.0% and it sets off a decline in all assets.

At the end of the cycle you have $100,000 and you have done well in PP.

1. Play that America is in a deep recession and the earning on the S&P 500 are at $55 and you apply a 7 P.E.  You get about 400 on the S&P 500 or a loss of maybe  68%.
2. Play that the long bond should be golden in this period, but rates are up.  You may have a loss of 50%. 
3. Should this happen first for whatever reason, it would likely drive us in to a deep recession much like what Volker did. 
4. Gold is your call.  Let’s say you lost 50% from where it was trading.  It was at $750 or so in 2008. 
5. Let’s play your cash is saving you and you still have 100% or $25,000.
Summary:
Stocks are worth $8,000
Long bonds are worth $12,500
Gold is worth $12,500
Cash is worth $25,000
Total=  $58,000
Loss of 42% and I declare you a BIG winner.

Why:

You have made money all the way to the end.  May be a long period before this plays out.

You have not lost any money until the end.

You have a big come back position.
You did not panic and sell at the bottom.

You know what you believe in: PP and rebalance.

You will not say you had the wrong Investments and start a new approach.

What if I can’t see right and PP lost less than 42%
If I can follow this script I will go to the beach and sit back for maybe 17 years or so and wait to rebalance PP.

[MediumTex]

That scenario above has some pretty wild projections in it.

It's very hard for me to imagine the stock market losing 68% and bonds also losing 50% of their value.  It's just hard to imagine money exiting the stock market and bond market at such a high rate.  Where would that money be going?

If fear was that high I'm not sure why gold would be losing 50% of its value.

I think that a more realistic tight money recession scenario is the one that actually occurred in the early 1980s where stocks did okay, cash did really well and bonds and gold did very poorly.

[moda0306]

MT,

Stocks lost -3.7% in 1981, while LT bonds netted out to 1.9% gain.

A tight money recession would more likely, IMO, be hard on stocks, as LT bonds act on investors' expectations of future rates, and don't necessarily move up as high as ST rates... hence the "inverted yield curve means a recession's coming" trend.

[MachineGhost]

This post and the one Clive made above it make sense to me. In a low interest rate environment, where is the downside?  Are there other factors in the decay besides borrowing costs for leverage?

Don't underestimate the power of tracking error to ruin you.  Check out this chart:



Green is spot.  Magenta 2x.  Turquoise 3x.

Unfortunately, the charting period is only limited to 3 months, but longer term the effects of decay and tracking error is absolutely devestating.

MG

[MediumTex]

I really don't understand why the decay illustrated above isn't showing up in the results of this portfolio.

The first time Wonk posted this idea I thought to myself "yeah, except the decay is going to kill you."

[clacy]

I really don't understand why the decay illustrated above isn't showing up in the results of this portfolio.

The first time Wonk posted this idea I thought to myself "yeah, except the decay is going to kill you."

MG's chart is for natty which will have contango effecting the ETF's which primarily rely on futures.

http://en.wikipedia.org/wiki/File:Conta ... dation.png

[MediumTex]

Great stuff Clive.  Thanks for posting.

[Lone Wolf]

Stocks lost -3.7% in 1981, while LT bonds netted out to 1.9% gain.

Wow, how'd that happen?  30-year rates were up something like 200 basis points in 1981.  I'd have thought that would bring on a large enough capital loss to surely keep LT's negative in '81.

Yet another factor to consider is that whilst you could shoot for the sky (20% annualised as this thread title suggests), its better IMO to instead only consider using leveraged funds to replicate a 1x PP.

Brilliant analysis as usual, Clive.  Great stuff.

[moda0306]

LW,

I'm using Craig's charts on PP returns, which arguably have softened losses during the 70's when the fund he uses as a measure isn't long-dated enough or something...

So I could be wrong on that.

[modeljc]

Lone Wolf,

Wow, how'd that happen?  30-year rates were up something like 200 basis points in 1981.  I'd have thought that would bring on a large enough capital loss to surely keep LT's negative in '81.

I don't remember the exact details but my boss retired with 100% of his money in tax free Munis.  After Volker raised rates he had 33% of his capital left to live out his life. 

[Lone Wolf]

2nd Jan 1981 30 year yield was 12.01% and on 31st December 1981 30 year yields were 13.65%

As no 29 year yield data is available, assuming the 29 year maturity yield was much the same as the 30 year then http://www.fixedincomeinvestor.co.uk/x/yieldcalc.html indicates a -11.75% capital loss and a 12.01% income = marginally positive overall total year gain (+0.26%)

Thanks, Clive!  I was drawing from what I remembered about the annual series.  (I looked at 1981 a while back when I was thinking about the viability of a Treasury Ladder riding out a 1981-like scenario.)

The "Annual" data has 1980 at 11.27% and 1981 at 13.45%, a spread of 218 basis points (versus the 164 basis points from the daily series.)  Negative performance for 1981, sure, but also not quite the disaster I'd have expected.  Interesting to see how nicely the high interest rates "buffer" the capital loss.

[MediumTex]

2nd Jan 1981 30 year yield was 12.01% and on 31st December 1981 30 year yields were 13.65%

As no 29 year yield data is available, assuming the 29 year maturity yield was much the same as the 30 year then http://www.fixedincomeinvestor.co.uk/x/yieldcalc.html indicates a -11.75% capital loss and a 12.01% income = marginally positive overall total year gain (+0.26%)

Thanks, Clive!  I was drawing from what I remembered about the annual series.  (I looked at 1981 a while back when I was thinking about the viability of a Treasury Ladder riding out a 1981-like scenario.)

The "Annual" data has 1980 at 11.27% and 1981 at 13.45%, a spread of 218 basis points (versus the 164 basis points from the daily series.)  Negative performance for 1981, sure, but also not quite the disaster I'd have expected.  Interesting to see how nicely the high interest rates "buffer" the capital loss.

Rates have been so low for so long I really haven't thought about how if interest rates were rising slowly you might still see positive returns (change in value + dividends) at various points on the yield curve.

[moda0306]

I can't imagine being around in 1981 to see both P/E ratios and bond yields at those rates.