Sequence of returns risk. Using puts to hedge.

[dandan14]

I'm considering an "early retirement" in the next year or so -- at least to the extent that I make a conscious decision to get off of the corporate merry-go-round and do something that likely pays a fraction of my current salary. Depending on what that is, we may or may not be able to live purely off of that new salary.

Wade Pfau and others have put a lot of thought into "sequence of returns" risk, which definitely applies in this scenario. One thing I was running the math on last night was potentially buying long term (LEAPS) puts against the S&P. If I buy puts at a strike price of 20% less than current price, the cost for a full year is about 2.1% of my portfolio value.

So my total potential loss would be 22.1% in any given year, and I'd pay 2.1% of potential gains for that insurance.

My other option, of course, is to shift my AA to something like 20/80 for the time being (5-6 years) until we have a better idea of what the next 20 years look like for us as a family.

Thoughts as it applies to early retirement and limiting risk?

[itstoomuch]

Thought about it
I do covered calls on individual stock and fortunately never collected. Have Always lost the option.
I'd look at GLWB deferred Variable annuities for some of your tax deferred. Start with Vanguard and then look at other vendors who have greater features.

Ymmv

[bigred77]

I think you would be much better off by just lowering your allocation to equities (although not all the way to 20/80, maybe something like 50/50).

Bogleheads obsess about lowering investment costs by just a few basis points and your strategy incurs 210 basis points of insurance cost. Plus the S&P may not be an appropriate hedging tool for you if you have a large chunk of international equities in your portfolio.

You also might find a Larry Swedroe type of portfolio designed to cut tail risk interesting. Also, not that i would recommend it, but you could explore a "costless collar" type strategy.

[CaliJim]

I think you would be much better off by just lowering your allocation to equities

+1

Also, if you are in Jim OTARs gray zone, consider an SPIA. viewtopic.php?t=144663

[lack_ey]

With puts that far out of the money, you're not really hedging all that much. Also, unless your asset allocation now is very conservative, the plan B of a switch all the way down to 20/80 would make much more of an impact, for better or worse.

With the puts, if the market goes -25%, great, you got a return of -22.1% instead? Next time you roll the hedge and buy some more puts, if projected vol is up then it'll cost more than that, so it's not like you pay this relatively small price indefinitely. In the next year it returns -17% and let's say you get a return of -20.5%. Awesome. You guarantee yourself the loss of premium but only noticeably do better than the market if it does really bad in a single year (as opposed to spread over multiple years).

Protective puts are generally bad risk/return. I know the start of early retirement is sensitive to downside risks and as such the option would be relatively more valuable than a more generalized analysis might suggest, but that's a high bar to clear for the pros to outweigh the cons. I really suspect that reducing market exposure is overall superior, though I've never done any simulations or any remotely serious analysis along those lines.

Now that I think of it, what was the extent of your "running the math" on this? Was it probabilistic or were you just looking at premiums, payouts, etc.?

[Rob5TCP]

One question - when do you by the LEAPS - for how many years out. What if the market is flat or relatively flat for a number of years. I did a couple of puts this summer (they will expire worthless) with the election on the horizon.

It is another form of market timing. As others have stated, you may have to lower your overall allocation so you can sleep at night.

[MN Finance]

I definitely understand the logic (and the counter points made), but the number one issue is one that was just mentioned by lackeys. You can't get a long term fixed insurance price based on today's market values. If you could then the discussion would be more interesting. In a multi year slowly declining market you keep rolling your options at a price 20% below market. You could realize a 50% decline over a few years and not be hedged (though options would increase in value, but would be hard to sell and take off protection at the right time)

[dandan14]

They do make LEAPS for 2+ years. The good news is that a 2 year leap on the S&P seems to be priced at the same 2.1% per year. So for about 4.2%, i could buy 20% downside protection for 2 years. So the "rolling risk" of 2 down years in a row would be insured as well.

[AlohaJoe]

Wade Pfau and others have put a lot of thought into "sequence of returns" risk, which definitely applies in this scenario. One thing I was running the math on last night was potentially buying long term (LEAPS) puts against the S&P. If I buy puts at a strike price of 20% less than current price, the cost for a full year is about 2.1% of my portfolio value.

So my total potential loss would be 22.1% in any given year, and I'd pay 2.1% of potential gains for that insurance.

FWIW, if you had that insurance it wouldn't have actually helped you any in the worst American historical retirement scenario. It probably helps to look into the historical failure scenario and understand why it failed:

If you had a 60/40 portfolio and retired in years 1965--1969 then you had the worst withdrawal rates in US history -- between 3.7% and 4.1%.

What caused those poor withdrawal rates wasn't some massive 40% market crash. In 1973 the market dropped 18% -- so your insurance wouldn't even have kicked in. And in 1974 the market dropped 27% so your insurance would have helped slightly but not enough.

But all those drops were in nominal terms. The real problems were caused by inflation. Here are the nominal stock market returns in the 1970s:



The (arithmetic) average over those years was 10.3%! Imagine if you told someone that 10.3% returns for a decade was bad. They'd think you were crazy. 40% of the time the annual return was over 20%.

Now look at it again in real terms -- with inflation adjustments.



When you adjust for inflation the average return drops to 2.58%.

If you look at other years with low safe withdrawal rates -- 1937 (4.18%) or 1912 (4.06%), say -- you see the same thing. If you retired in 1937 you were hit by 11% inflation in 1941, 7% in 1942, 18% in 1946, 10% in 1947. If you retired in 1912 you saw 12% inflation in 1916, 19% in 1917, 17% in 1918, 16% in 1919.

Nominal returns weren't the problem; inflation was.

So if nominal returns weren't the problem, is buying insurance for poor nominal returns the answer?

Sequence of returns risk is about sequence of real returns and those problems are (at least in the US historical record) always caused by inflation, not by poor equity returns.

You shouldn't fear a stock market crash in the early part of your retirement. You should fear high inflation that turns decent nominal returns into poor real returns.

(Disclaimer: other countries have had other problems and the future is unlikely to exactly repeat the past. Still, drawing the wrong conclusions from the historical record is still wrong )

[lack_ey]

By the way, if you're set on this, consider breaking up the hedge into multiple tiers.

For example, buy puts for a third the stock allocation value at -10%, another third at -20%, and the rest at -30%.

Again, I haven't looked into this in any real capacity, but at least this is what my intuition is telling me: if you don't have a known liability to match or reason to hedge some exact amount like to -20% specifically, you may be better off making the bet less discrete. Not like that will make some huge difference in the outcome, but I don't immediately see why not.

P.S. backing up, I still would need some convincing that this is a good idea in the first place

[longinvest]

Just a little reminder that "sequence of returns risk" is very easy to avoid, in retirement. One can simply use a flexible withdrawal method, instead of the 4% SWR which was never meant as a withdrawal method in the first place! (SWR is a way to calculate how much one needs to retire at 65).

Unfortunately, eliminating "sequence of returns risk" does not eliminate "market risk", to which flexible withdrawal methods are exposed.

This post by Skjoldur explains the difference between "sequence of returns risk" and "market risk":

viewtopic.php?f=2&t=190721&start=50#p2900256

Maybe I can help illustrate the difference between what longinvest refers to as 'market risk' and sequence of returns risk. I have made three graphs using made up numbers.

I made 10 random returns, and then scrambled those 10 returns into a variety of different sequences. The first picture shows a $1M portfolio following each of these different paths. Note that the final portfolio value is identical in each case. Holding a portfolio without adding or withdrawing is not subject to sequence of returns risk. The sequence does not matter.



The next image shows the same portfolio with a 4% annual withdrawal. This is not a constant dollar withdrawal, it is a constant percentage withdrawal. Note, that the end values of all the paths are the same. Constant percentage withdrawal is not subject to sequences of returns risk. The portfolios are subject to market risk, in that the ending value could be high or low but the sequence does not matter.



The final image shows the same portfolio with a $40K withdrawal, this is the dreaded constant dollar withdrawal. Note that in this case, the final portfolio values are all different. That difference is sequence of returns risk. In this case, poor returns early combined with constant withdrawals results in varied outcomes.



It turns out (and this surprised me mathematically, but longinvest demonstrated it another thread) that a sequence of varied percentage withdrawals has the same property as a constant percentage withdrawal. In other words, VPW is also mathematically immune to sequence of returns risk.

So here is a bonus picture. In this one, the portfolios are all subject to the same sequence of varied percentage withdrawals. You can see that the end portfolio values are all the same once again.



So VPW is immune to sequence of returns risk with regard to final portfolio value.

That does not mean that VPW is not 'risky' or that it magically fixes the fact that the markets themselves are 'risky.' But (along with similar percentage based withdrawal methods) it does not suffer from sequence of returns risk. How cool is that!

[itstoomuch]

By the way, if you're set on this, consider breaking up the hedge into multiple tiers.

For example, buy puts for a third the stock allocation value at -10%, another third at -20%, and the rest at -30%.

Again, I haven't looked into this in any real capacity, but at least this is what my intuition is telling me: if you don't have a known liability to match or reason to hedge some exact amount like to -20% specifically, you may be better off making the bet less discrete. Not like that will make some huge difference in the outcome, but I don't immediately see why not.

P.S. backing up, I still would need some convincing that this is a good idea in the first place

I figure that all my mechanizations, I only need to beat only some of the BHers participating on this thread.

YPuts&CallsMV

[Watty]

One thing I was running the math on last night was potentially buying long term (LEAPS) puts against the S&P.

One problem is that a large percentage of your portfolio will be in bonds and that would not hedge a bear market in bonds. You also need to consider your bond return in real inflation adjusted dollars.

Depending on what that is, we may or may not be able to live purely off of that new salary.

You would need to play with the numbers but if the gap is something like $10K a year for ten years then you could just make a ladder of CDs or government bonds to cover that.

If you would not need to make any withdrawals from your portfolio for ten years then you would not need to worry about the sequence of returns risk.

You might also look at way to reduce your expenses, like paying off or paying down a mortgage, to eliminate the gap.

[itstoomuch]

Break up the options into tiers and time.
Y"options"MV

I find Withdrawal Strategy a lot more challenging than Putting it in.
:LOL:

[RNJ]

Depending on your age, couldn't you buy a TIPS ladder of the # of years you want to hedge?

[btenny]

Go look into what happened in 1987 fast market crash for those with "portfolio insurance". People were doing things similar to what you suggest. I do not know the details. It did not work out well.

https://en.wikipedia.org/wiki/Portfolio_insurance

[dandan14]

Wade Pfau and others have put a lot of thought into "sequence of returns" risk, which definitely applies in this scenario. One thing I was running the math on last night was potentially buying long term (LEAPS) puts against the S&P. If I buy puts at a strike price of 20% less than current price, the cost for a full year is about 2.1% of my portfolio value.

So my total potential loss would be 22.1% in any given year, and I'd pay 2.1% of potential gains for that insurance.

FWIW, if you had that insurance it wouldn't have actually helped you any in the worst American historical retirement scenario. It probably helps to look into the historical failure scenario and understand why it failed:

If you had a 60/40 portfolio and retired in years 1965--1969 then you had the worst withdrawal rates in US history -- between 3.7% and 4.1%.

What caused those poor withdrawal rates wasn't some massive 40% market crash. In 1973 the market dropped 18% -- so your insurance wouldn't even have kicked in. And in 1974 the market dropped 27% so your insurance would have helped slightly but not enough.

But all those drops were in nominal terms. The real problems were caused by inflation. Here are the nominal stock market returns in the 1970s:



The (arithmetic) average over those years was 10.3%! Imagine if you told someone that 10.3% returns for a decade was bad. They'd think you were crazy. 40% of the time the annual return was over 20%.

Now look at it again in real terms -- with inflation adjustments.



When you adjust for inflation the average return drops to 2.58%.

If you look at other years with low safe withdrawal rates -- 1937 (4.18%) or 1912 (4.06%), say -- you see the same thing. If you retired in 1937 you were hit by 11% inflation in 1941, 7% in 1942, 18% in 1946, 10% in 1947. If you retired in 1912 you saw 12% inflation in 1916, 19% in 1917, 17% in 1918, 16% in 1919.

Nominal returns weren't the problem; inflation was.

So if nominal returns weren't the problem, is buying insurance for poor nominal returns the answer?

Sequence of returns risk is about sequence of real returns and those problems are (at least in the US historical record) always caused by inflation, not by poor equity returns.

You shouldn't fear a stock market crash in the early part of your retirement. You should fear high inflation that turns decent nominal returns into poor real returns.

(Disclaimer: other countries have had other problems and the future is unlikely to exactly repeat the past. Still, drawing the wrong conclusions from the historical record is still wrong )

Excellent points.

[lack_ey]

Go look into what happened in 1987 fast market crash for those with "portfolio insurance". People were doing things similar to what you suggest. I do not know the details. It did not work out well.

https://en.wikipedia.org/wiki/Portfolio_insurance

The mechanics are different. The 1987 crash, which was caused by a number of factors, is not any kind of relevant example or cautionary tale about the behavior of puts purchased far in advance. In 1987 there was a lot of money invested with certain concepts and strategies, a lot of selling in the futures markets, a positive feedback loop that developed, and people attempting to exploit others to reinforce the loop. The act of buying puts beforehand or exercising them doesn't create that kind of immediate pricing pressure on the markets during the time of stress.

[hirlaw]

I am 60 and very close (as in probably within one year) to retirement. To mitigate sequence of return risk, over the past 2 years I have gradually cut back my equity portion from about 55% to its present 38%. My goal is to have it to around 35% at retirement. After dividends/interest, retirement spending amounts will be taken primarily from the fixed income portion. So, barring a continued bull market, it will probably be an upward sloping glidepath until the portfolio reaches about 50-50 -- at which point I will more than likely keep it steady.

[Clive]

One thing I was running the math on last night was potentially buying long term (LEAPS) puts against the S&P. If I buy puts at a strike price of 20% less than current price, the cost for a full year is about 2.1% of my portfolio value.

So my total potential loss would be 22.1% in any given year, and I'd pay 2.1% of potential gains for that insurance.

My other option, of course, is to shift my AA to something like 20/80 for the time being (5-6 years) until we have a better idea of what the next 20 years look like for us as a family.

A good form of sequence of return risk reduction is to lower your spending and/or postpone retirement (save more). There is marked differences even for 0.1% differences in SWR. 0.5% can make a massive difference. If you can scale down initial SWR with a view to perhaps uplifting that at a later date if things are ticking along nicely then that will be more protective. If you can also save/accumulate a little longer before retiring then you might move from 25 times (4% SWR) from perhaps a portfolio of say $1M ($40K income) ... to 33 times - perhaps $1.2M portfolio and 3% SWR ($36K income).

Another choice instead of writing puts is to follow Zvi Bodie style. He allocates something like 10% to LEAP's and considers them to be 10x leveraged type holdings, alongside 90% in safe (TIPS in his case). That can reflect 100% unleveraged long, but has the added benefit that between rebalance points the maximum downside is less than 10% (less than 5% if you target 50/50 long/safe i.e. 5% LEAPS). The risk then is repeated adding to losing LEAPs sequentially too quickly, however if the rebalance interval is 12 months then your single year downside is potentially small.

Variations of that is the likes of XIV (sell volatility) that is broadly comparable to 5x. Or even a 3x or 2x stock ETF might be used. For example 2x SSO, weighted to 50/50 stock/bond desired exposure = 25% SSO, 75% BND

Or using XIV (shorter history, best viewed by selecting the LOG plot option for the second to last chart)

Note however, especially in the first of the above two links, that the apparent safety can be a illusion and comparable drawdowns can still arise. Similar for buying PUT LEAPS, the cost (you suggest 2.1% in your case) might amount to the risk premium/reward, such that overall its little different to just having invested in cash deposits on a broad risk/adjusted basis.

[David Jay]

Dan:

One data point. Using historic performance since 1926, the portfolio yield difference between a 100% stock portfolio and a 50/50 stock/bond portfolio is 1.9% (10.2% versus 8.3%), less than the 2.1% cost of your strategy.

The worst year for a 50/50 portfolio was 1931, with a return of -22.5%, within a fraction of your protection number of 22.1%.

data source: https://personal.vanguard.com/us/insigh ... llocations

[hirlaw]

The worst year for a 50/50 portfolio was 1931, with a return of -22.5%, within a fraction of your protection number of 22.1%.

data source: https://personal.vanguard.com/us/insigh ... llocations

Slightly off topic but I never know how to deal with that Vanguard chart because it is in calendar years. For example, under the chart, the worst calendar year performance for stocks was -43.1% in 1931. However, the chart doesn't deal with rolling 12 month periods. The worst 12-month period for U.S. stocks (the 12 months ended June 1932) was –67.6%.

[bigred77]

Dan:

One data point. Using historic performance since 1926, the portfolio yield difference between a 100% stock portfolio and a 50/50 stock/bond portfolio is 1.9% (10.2% versus 8.3%), less than the 2.1% cost of your strategy.

The worst year for a 50/50 portfolio was 1931, with a return of -22.5%, within a fraction of your protection number of 22.1%.

data source: https://personal.vanguard.com/us/insigh ... llocations

That's pretty compelling to me.

I'd assume that adding some international diversification to the equity side would reduce the worst year draw down as well.

[Dutch]

Just a little reminder that "sequence of returns risk" is very easy to avoid, in retirement. One can simply use a flexible withdrawal method, instead of the 4% SWR which was never meant as a withdrawal method in the first place! (SWR is a way to calculate how much one needs to retire at 65).

Unfortunately, eliminating "sequence of returns risk" does not eliminate "market risk", to which flexible withdrawal methods are exposed.

I've seen you state this before, and I don't think it's particularly helpful. In fact, I think you're just confusing people.

A market downturn/crash at the beginning of retirement will be a problem regardless of withdrawal method.

- A fixed withdrawal method will have you run the risk of prematurely running out of money
- A variable withdrawal method avoids this risk by dramatically cutting back on the annual withdrawals. So, at the expense of the standard of living.

Either one is a problem, no matter what name you choose to give it.

BTW Thanks for all your work on VPW. I'm a big fan.

[longinvest]

Just a little reminder that "sequence of returns risk" is very easy to avoid, in retirement. One can simply use a flexible withdrawal method, instead of the 4% SWR which was never meant as a withdrawal method in the first place! (SWR is a way to calculate how much one needs to retire at 65).

Unfortunately, eliminating "sequence of returns risk" does not eliminate "market risk", to which flexible withdrawal methods are exposed.

I've seen you state this before, and I don't think it's particularly helpful. In fact, I think you're just confusing people.

A market downturn/crash at the beginning of retirement will be a problem regardless of withdrawal method.

- A fixed withdrawal method will have you run the risk of prematurely running out of money
- A variable withdrawal method avoids this risk by dramatically cutting back on the annual withdrawals. So, at the expense of the standard of living.

Either one is a problem, no matter what name you choose to give it.

BTW Thanks for all your work on VPW. I'm a big fan.

Dear Dutch,

Sequence of return risk is a risk of bankruptcy. In other words, it's the risk that one hits a bad patch at the start of retirement and ends up bankrupt 20 years later, long after the crisis, because the portfolio couldn't recover. That means no portfolio left. Withdrawals are cut to zero. This is a fatal risk for the portfolio, even though it has a small probability of happening.

Variable withdrawals, on the other hand, reduce withdrawals at the moment it counts, during a bad crisis, so that the portfolio survives and has a chance to recover when better times come. The market risk of variable withdrawals is about cutting expenses when markets are bad, workers are losing their job, and so on. It's traveling within the US instead of abroad, dining out a little less, and so on, during retirement. That's a much smaller risk*.

* I mean that the financial consequences of hitting a bad patch are much smaller.

I don't think that I am confusing people.

Regards,

longinvest

[hirlaw]

Not sure that simply varying a withdrawal rate works.

65 year old retiree has $1 million in an all equity portfolio. He is taking 4% a year withdrawals. ($40k for the relevant year). Market drops 60%. He now has $400k. Even cutting his spending in half, from $40k to $20k, brings his withdrawal rate up to 5%. It will be VERY difficult for that retiree to recover.

[longinvest]

In our wiki:

https://www.bogleheads.org/wiki/Variabl ... withdrawal

Variable percentage withdrawal (VPW) is a withdrawal method that adapts to the retiree's retirement horizon, asset allocation, and portfolio returns during retirement. It combines the best ideas of the constant-dollar, constant-percentage, and 1/N withdrawal methods to allow the retiree to spend most of his portfolio using return-adjusted withdrawals. By adapting withdrawals to market returns, VPW will never prematurely deplete the portfolio.
...

How to use variable-percentage withdrawals during retirement

VPW is best used in conjunction with guaranteed base income from Social Security, pensions, and, if necessary, inflation-indexed Single Premium Immediate Annuity (SPIA).

Base income for years between retirement and the start of Social Security payments can be provided by using a simple CD ladder. For the purposes of VPW calculations, the money set aside in the CD ladder should not be considered as part of the portfolio.

It has been suggested to delay Social Security payments to increase base income in Bogleheads forum topic: Delay Social Security to age 70 and Spend more money at 62.
...

(I underlined part of the text.)