Canadian Money Forum banner

61 - 78 of 78 Posts

·
Registered
Joined
·
17,027 Posts
Discussion Starter #61
I don't think SH will compare to TAIL, because by design SH should negate any movement of SPY on a daily basis. Holding both long term, would cancel SPY's returns and add a decay component from SH to it, resulting in a gradual loss.
Good point. And the daily tracking nature of SH pretty much guarantees decay over time. TAIL doesn't have that problem.

Combining SPY with TAIL would be like buy an ITM call option, with the added optionality of being able to decouple the trade at opportune occasions.
Makes sense, but I don't know anything about modelling those combinations (I don't know much about options).

I attempted to walk through a monthly simulation to see what happened with a SPY/TAIL mix with monthly rebalancing. Along the lines of your earlier suggestion.

First just SPY alone. From 2018-07-01 to 2020-06-01, the total return including dividends is 8.3% CAGR. The worst drawdown using monthly data was -24%.
  • 70% SPY 30% TAIL. Now the return is 9.0% CAGR with -10% drawdown... better!
  • 50% SPY 50% TAIL. Now the return is 8.4% CAGR with -2% drawdown. Just wacky.
Posting the chart separately below
 

·
Registered
Joined
·
17,027 Posts
Discussion Starter #62
It should be noted that the 70/30 mix underperformed much of the time; notice that the blue line stays below the orange line until COVID-19.

But, the end result over these ~ 2 years is that the 70/30 actually beat SPY, with far less volatility. In other words, far superior risk adjusted return (Sharpe & Sortino). Monthly rebalancing is required to achieve this magic.

20263
 

·
Registered
Joined
·
17,027 Posts
Discussion Starter #63
If one does a 50/50 with stocks, the portfolio will be protected from severe losses (and no need for bonds), but overall returns could suffer if there are losses on the TAIL side. On the other hand, one could lever it up, given the downside is limited. The rebalancing bonus could be substantial.
My simulation agrees with your intuition here. Starting from 2018-07-01, I'm seeing the 50/50 mix give the same CAGR as 100% stocks. I'm rebalancing monthly. Volatility has been virtually eliminated with only 1.5% drawdown during the recent crash.

Graph not shown for this one. See previous post for 70/30.

What a stunning result. Same performance as 100% stocks, with no volatility. Mechanical and no market timing required.

Could it act like this in the long term? What would happen during a long and steady stock rally?
 

·
Registered
Joined
·
851 Posts
...though buying new put option for the roll over when premiums were high would put a drag on the fund.
I totally agree with this. There is no point in buying puts when the VIX is in the 80s. If anything, that would be a good time to go short. One way to do that using the ETF is to sell all of the units and use the proceeds to buy SPY for example. For adventurous traders, there is the choice of shorting TAIL above a certain levels of VIX.
 

·
Registered
Joined
·
851 Posts
But, the end result over these ~ 2 years is that the 70/30 actually beat SPY, with far less volatility.
The last 2 years have seen two bouts of high volatility, one at the end of 2018 and another one recently. But volatility was generally muted in the past decade. Would the combo suffer from chronic under-performance under those conditions?
 

·
Registered
Joined
·
851 Posts
My simulation agrees with your intuition here. Starting from 2018-07-01, I'm seeing the 50/50 mix give the same CAGR as 100% stocks. I'm rebalancing monthly. Volatility has been virtually eliminated with only 1.5% drawdown during the recent crash.
Not sure I read the chart the same. I think there was a 10% draw down, from 116k to 104k, but it is an improvement over SPY nonetheless.

If volatility, particularly extreme volatility, could be eliminated from the combination, it would be ripe for leverage to boost returns.
 

·
Registered
Joined
·
17,027 Posts
Discussion Starter #67
The last 2 years have seen two bouts of high volatility, one at the end of 2018 and another one recently. But volatility was generally muted in the past decade. Would the combo suffer from chronic under-performance under those conditions?
That's a good point. We've had nice volatility in the last 2 years. That means index puts (and TAIL) have soared at times, which in turn has created the rebalancing opportunity that is harvested in my simulations above.

In contrast, a calm perpetual bull market would not create those volatility spikes. TAIL would chronically perform poorly and there is no rebalancing advantage. And I think you can see a hint of this in the graph I posted, in 2019. If the market behaved like 2019, for years on end, the returns of this combo would be poor.

Not sure I read the chart the same. I think there was a 10% draw down, from 116k to 104k, but it is an improvement over SPY nonetheless.
The chart I posted was 70/30 and had 10% draw down, as you say. But I did not show the chart of 50/50, which according to my paper simulation had just 2% draw down.

If volatility, particularly extreme volatility, could be eliminated from the combination, it would be ripe for leverage to boost returns.
This may be possible with a 50/50 combo.
 

·
Registered
Joined
·
851 Posts
The chart I posted was 70/30 and had 10% draw down, as you say. But I did not show the chart of 50/50, which according to my paper simulation had just 2% draw down.
Sorry, my bad.

A 2% draw down is virtually nothing. When long options go in the money, they gain deltas, so I wouldn't be surprised if a 50% SPY drop would have given the combination a small upside potential. Maybe 50/50 is a bit too much insurance and 60/40 is more in line with a reasonably protected portfolio.

With regards to using margin, it is the deep and fast drops that create problems of margin restriction. TAIL is likely to do a good job to protect against those. Of course if it restricts the upside too much, then leverage would not be as effective. Better do the good old stocks and bonds.
 

·
Registered
Joined
·
17,027 Posts
Discussion Starter #69 (Edited)
With regards to using margin, it is the deep and fast drops that create problems of margin restriction. TAIL is likely to do a good job to protect against those. Of course if it restricts the upside too much, then leverage would not be as effective. Better do the good old stocks and bonds.
I agree that, theoretically, the SPY+TAIL combo looks promising somewhere like IB where you can get cheap margin rates and cheap trades (aggressive rebalancing). I would agree that it should protect you from those sharp drops and margin problems, unless IB changes broader margin policies. I wonder how they treat TAIL for example.

Hedging is a tough balance though. Add too much insurance and you'll get a mild portfolio, but won't perform so well. Or you may end up jumping through a lot of hoops (many trades and margin complexities) only to end up with something similar to a more conventional diversified portfolio.

As a quick example, I was on a Cuban beach during the worst part of the recent stock market crash. It was not possible to log into a brokerage and place trades, and yet, with the SPY+TAIL combo that's probably when you'd want to do a rebalancing to harvest a huge movement. By the time I got back to Canada, the market had calmed down quite a bit.

This actually is not the first time the stock market crashed on me while on a beach somewhere. For me, the passive portfolios have a big advantage since they "do their thing" while unattended.

Similarly, if using IB margin, you definitely need to stay on top of that portfolio.
 

·
Registered
Joined
·
17,027 Posts
Discussion Starter #70
Maybe 50/50 is a bit too much insurance and 60/40 is more in line with a reasonably protected portfolio.
Let's try this 60% SPY, 40% TAIL mix. I'm also curious what happens if you don't rebalance along the way:

YTD return on June 22 is +4.6% vs -2.5% for SPY
The low point was -6.4% vs -30.3% for SPY (Dec 31 - Mar 23)

That looks like pretty substantial protection / reduction in volatility to me, even with no rebalancing. Softening that 30% drop down to just 6% would be a huge win. The only caveat I see is that the chosen time period happened to start with rock bottom VIX and therefore a great entry to TAIL.

What do you think, Topo?

Code:
Starting values December 31: SPY=60,000, TAIL=40,000, Total=100,000
February 28 update: SPY=55,230, TAIL=44,452 , Total=99,682
March 23 update (market low): SPY=41,808, TAIL=51,828, Total=93,636
April 30 update: SPY=54,468, TAIL=47,240, Total=101,708
June 22 update: SPY=58,500, TAIL=46,076, Total=104,576
 

·
Registered
Joined
·
851 Posts
Let's try this 60% SPY, 40% TAIL mix. I'm also curious what happens if you don't rebalance along the way:

YTD return on June 22 is +4.6% vs -2.5% for SPY
The low point was -6.4% vs -30.3% for SPY (Dec 31 - Mar 23)

That looks like pretty substantial protection / reduction in volatility to me, even with no rebalancing. Softening that 30% drop down to just 6% would be a huge win. The only caveat I see is that the chosen time period happened to start with rock bottom VIX and therefore a great entry to TAIL.

What do you think, Topo?

Code:
Starting values December 31: SPY=60,000, TAIL=40,000, Total=100,000
February 28 update: SPY=55,230, TAIL=44,452 , Total=99,682
March 23 update (market low): SPY=41,808, TAIL=51,828, Total=93,636
April 30 update: SPY=54,468, TAIL=47,240, Total=101,708
June 22 update: SPY=58,500, TAIL=46,076, Total=104,576
Certainly interesting. What I am curious about is for example in a 50% SPY drop, how much would TAIL pop? With a 30% drop in SPY, TAIL gained almost 25%. Given there are options in the portfolio that would gain deltas and the volatility would explode to the upside, I would not be surprised if TAIL will outdo SPY and a 50/50 would actually gain value in the midst of a bear market! This assumes TAIL has not lost a big chunk to decay in the months prior. But even then rebalancing would have partly replenished the TAIL side and kept it ready for the pop. We of course do not know and have to wait to see how it performs.
 

·
Registered
Joined
·
17,027 Posts
Discussion Starter #72
Certainly interesting. What I am curious about is for example in a 50% SPY drop, how much would TAIL pop? With a 30% drop in SPY, TAIL gained almost 25%. Given there are options in the portfolio that would gain deltas and the volatility would explode to the upside, I would not be surprised if TAIL will outdo SPY and a 50/50 would actually gain value in the midst of a bear market!
I agree, I think the mix would do great during a big drop.

This assumes TAIL has not lost a big chunk to decay in the months prior. But even then rebalancing would have partly replenished the TAIL side and kept it ready for the pop.
It's the performance during low VIX market rallies that worries me. Look at 2017-05-01 to 2018-10-01, when TAIL lost 18%.

That's a pretty bad drag over a lengthy period if mixed with SPY. Would that be a problem?

Also, I wonder how often you'd have to rebalance for best effect.
 

·
Registered
Joined
·
17,027 Posts
Discussion Starter #73
I just adjusted my US retirement plan, where I have no trade fees. Previously, my stock component was entirely in SPY.

I changed this to 80% SPY + 20% TAIL to get some downside protection.

It's a minor change, but the protective puts will partially protect my SPY position and I want to see how it responds over a few months.
 

·
Registered
Joined
·
851 Posts
It's the performance during low VIX market rallies that worries me. Look at 2017-05-01 to 2018-10-01, when TAIL lost 18%.

That's a pretty bad drag over a lengthy period if mixed with SPY. Would that be a problem?
It could be. It would be similar to buying SPY with protective puts. In a bull or neutral market, SPY alone will do better than the combo.

Also, I wonder how often you'd have to rebalance for best effect.
I definitely would aggressively rebalance from TAIL to SPY, particularly when VIX is high and the market has fallen more than 10 percent. From SPY to TAIL, I would rebalance maybe once every 6 months or so, since volatility tends to spike about twice per year. I would not find it easy to rebalance repeatedly into a "sagging TAIL."

One could start with a lighter allocation to TAIL, such as the 20% you have chosen, to leave more room for rebalancing. But I don't know what the optimum strategy would be. If one had invested since inception without rebalancing, there would have been only 2 occasions that one could break even. Other times the trade was a loser.

I hope the chart does not end up looking like UVXY. Time will tell.
 

·
Registered
Joined
·
17,027 Posts
Discussion Starter #75
I definitely would aggressively rebalance from TAIL to SPY, particularly when VIX is high and the market has fallen more than 10 percent. From SPY to TAIL, I would rebalance maybe once every 6 months or so, since volatility tends to spike about twice per year. I would not find it easy to rebalance repeatedly into a "sagging TAIL."
Thanks, that sounds like a very good plan. You've even accounted for the behavioural challenge of rebalancing from SPY to TAIL.

I hope the chart does not end up looking like UVXY. Time will tell.
One major difference is that UVXY is both leveraged and rebalances (or somehow resets) daily. Most of those "daily" tracking funds, especially the leveraged ones, seem to develop this brutal decay. I'm hoping that TAIL is different since it is a portfolio of put options, held much like any other portfolio of securities. They don't track a daily performance.
 

·
Registered
Joined
·
336 Posts
My portfolio is simple. Just a handful of ETFs. One ETF (XIC) for Canadian equity. One ETF (XAW) for American+international equity. One ETF for bonds (XBB). One ETF for REITs. One ETF for gold. Then some cash in a HISA.

That's it. Pretty simple, I think. I try not to time the market, and I don't do anything fancy.

The trick is to get the allocation between these different ETFs right for me. I'm fairly conservative and keep about 55% my money in cash+gold+bonds, and the rest in stocks + REITs.
 

·
Registered
Joined
·
17,027 Posts
Discussion Starter #77
My portfolio is simple. Just a handful of ETFs. One ETF (XIC) for Canadian equity. One ETF (XAW) for American+international equity. One ETF for bonds (XBB). One ETF for REITs. One ETF for gold. Then some cash in a HISA.

That's it. Pretty simple, I think. I try not to time the market, and I don't do anything fancy.
Not doing anything fancy has some major benefits. Whenever one tries to do anything fancy, there's the risk of causing harm (making things worse than before). Staying totally passive, just buying and holding the ETF mix as you do, minimizes the number of things that can go wrong.

It's also much less work. Though I'm intrigued about this hedging of the S&P 500, it would involve a lot of monitoring and manual trades over time. That takes up time and energy.

Also, by holding stocks & bonds & gold too, you're already getting quite a bit of hedging -- without having to do anything fancy.
 

·
Registered
Joined
·
17,027 Posts
Discussion Starter #78
Now (near market highs) is the time to think about risk mitigation.

Ritholtz and Meb Faber share some observations in this article about timing the market.

I apply a small amount of market timing (similar to the methods they describe) for 16% of my equities. I guess it's 20% once you add in the TAIL strategy I recently started. The remaining 80% of my equities are unprotected, always long.

Meb said something interesting in another interview I heard from him: These kinds of technical methods for timing the market (getting out) are probably not going to increase your return. The reason for using them is to defend yourself against the really big drops, like 50%, 60%, even 80% decline in the index... which has happened in the US and elsewhere in the world. Sometimes the market stays depressed for many years.

I agree with Meb that protecting oneself against those catastrophic crashes is worthwhile. It would be very unpleasant to experience a 70% equity loss which doesn't recover for many years. This kind of protection likely does have a "cost" in the form of lost performance... and I'm OK with that.

I'm curious, who else here uses some kind of technical cues for getting out of stocks in case we have a severe crash? This seems like a good time to think about this, as we may be in the early stages of a true Depression.
 
61 - 78 of 78 Posts
Top