Author Topic: bonds as retirement date insurance--premium too high  (Read 921 times)

Roothy

  • Stubble
  • **
  • Posts: 117
bonds as retirement date insurance--premium too high
« on: June 11, 2018, 05:55:37 PM »
I am about a year and a half from retirement (I hope), and am at about 90/10 stocks/bonds.  I know I "should" be moving more into bonds to protect my retirement date (as I would simply keep working if stocks tanked before I reached my "number"/date... something that would dismay me a lot).

To make myself willing to do it, I decided to think of it as buying retirement date insurance.  But once I did that, I realized the premium for such insurance is really, really high.  To "protect" $100,000 by moving it from stocks to bonds, you're paying about $4000 in opportunity costs/likely foregone gains.  That is, assuming $100,000 in stocks would earn $7000 in an average year, and bonds would earn more like $3000, the net "loss" from moving to bonds is $4000/year.  And that's only to protect $100,000! 

Once I thought about it this way, it's getting harder to stuff the genie back into the bottle, cognitively.  I can't bear the idea of "paying" a $4000 premium every year (really, more like $32,000 given the asset allocation that the experts recommend you enter retirement with), just to avoid the possibility of having to work an extra year or so in the event of a big market dump right before I retire.

I know I'm thinking about this wrong.  Help!
« Last Edit: June 11, 2018, 06:42:35 PM by Roothy »

Roothy

  • Stubble
  • **
  • Posts: 117
Re: bonds as retirement date insurance--premium too high
« Reply #1 on: June 11, 2018, 06:45:13 PM »
And now that I'm thinking about it, it's probably even worse than that.  Worst-case (but still reasonably realistic) scenario, the market would drop 50%.  That means I'd lose $50,000 on $100,000 in stocks.  If instead I moved that $100,000 into bonds to protect it, the $4000 I'd be "paying" in a "premium" would realistically only be preserving $50,000 in thwarted losses.  $4000 to protect $50,000--each and every year I keep that allocation--seems insane to me.  I'm sure I'm missing something.

Radagast

  • Handlebar Stache
  • *****
  • Posts: 1189
  • Location: West of the Mountains, East of the Sea
Re: bonds as retirement date insurance--premium too high
« Reply #2 on: June 11, 2018, 08:26:52 PM »
I think you have it mostly right. I assume you know about the bond tent or reverse equity glide path? (Seach for it if not for a better explanation). Basically start buying CDs right now, and by the time you quit have 5 years of CDs so that when you quit you will not sell any stocks for 5 years and in fact will keep reinvesting the dividends the whole time as you live on maturing CDs.

Or gradually switch your bond allocation to 40% by the time you quit, and reinvest dividends for the first five years while living off bonds, then live off bonds and dividends for another five years, and only after 10 years resume your 90/10 porfolio selling whatever when you need cash.

Historically this reduced the best and probably even the average outcomes in exchange for avoiding the worst. That takes away a little of the gambling element, but you are correct that it is insurance that may never pay for itself.

Goldielocks

  • Walrus Stache
  • *******
  • Posts: 5591
  • Location: BC
Re: bonds as retirement date insurance--premium too high
« Reply #3 on: June 11, 2018, 09:38:07 PM »
I recommend taking a look at your MINIMUM expenses that you need to withdraw from your accounts after you pass an "age of no return to work". By minimum, I mean to feed yourself, heat your home and pay for rent/mortgage/taxes.

e.g., age 65,you decide you need $10k/yr to top up your SS payments.

Then, you start a bond ladder to generate $10k/yr starting at age 65.  Yes, it is low return, but it is guaranteed, like buying your own pension, and you could sell bond in a severe need (might lose money).

Then you can be higher risk with the (larger amount) of the remainder.

Different people have different amounts, but I truly think that locking in your MINIMUM before early retirement, into a guaranteed place, is a great idea.

Nate79

  • 5 O'Clock Shadow
  • *
  • Posts: 80
Re: bonds as retirement date insurance--premium too high
« Reply #4 on: June 11, 2018, 11:21:53 PM »
The problem is not just that the market can drop 50% (by the way, that is not the worst case scenario). The problem is that in retirement you are also drawing down the account at the same time. The market can take years to recover, all the time you are withdrawing money to live on. The market can take anywhere from 5-10 years to recover in the worst case.

Sent from my SM-G950U using Tapatalk


DreamFIRE

  • Handlebar Stache
  • *****
  • Posts: 1291

Systems101

  • Stubble
  • **
  • Posts: 149
Re: bonds as retirement date insurance--premium too high
« Reply #6 on: June 12, 2018, 07:58:26 AM »
I can't bear the idea of "paying" a $4000 premium every year (really, more like $32,000 given the asset allocation that the experts recommend you enter retirement with), just to avoid the possibility of having to work an extra year or so in the event of a big market dump right before I retire.

That's pretty much what insurance is... it provides protection against a possible situation, at the cost of an ongoing premium.  In general, it's crazy profitable for the insurer, which is why at some point, you self-insure.  For example, I no longer have collision coverage on my car.  The 10%+ of vehicle value per year is a silly premium to pay.... but I also have enough $ that paying cash for a new(er) car wouldn't be an issue.

The question is whether you can *afford* to self-insure, and to do that, you don't look at the average statistics (which is where you are drawing the $4K)... you look at the downside, in the situation where the event occurs... do you *think* it's one more year, or have you actually done analysis to show it?  What if it's 2 ... or 4?

I echo Radagast and strongly suggest you look through the analysis folks have done on the reverse equity glide path - regardless of what direction you choose, I suspect it will help you make the trade-off you are facing.


MustacheAndaHalf

  • Handlebar Stache
  • *****
  • Posts: 1254
Re: bonds as retirement date insurance--premium too high
« Reply #7 on: June 12, 2018, 08:50:29 AM »
That is, assuming $100,000 in stocks would earn $7000 in an average year, and bonds would earn more like $3000, the net "loss" from moving to bonds is $4000/year.
Stock returns look more like 5% +/- 20% than your fixed 7%.  A fixed 7% looks more like bonds than stocks.

You're also ignoring your needs.  If you have enough to retire, what does doubling that do?  You can't retire twice over.  But you can lose enough to force you back to work.  If you have enough, why do you need to take additional risks?

talltexan

  • Handlebar Stache
  • *****
  • Posts: 1930
Re: bonds as retirement date insurance--premium too high
« Reply #8 on: June 12, 2018, 10:05:15 AM »
That difference in return is given to stocks precisely because they're so much more volatile in the short-term. Your 90/10 mix means you would have 2.5 years of expenses in safe instruments at the time of retirement. I'd look in the data about how often it takes the SP500 more than 2.5 years to recover to a pre-bear-market peak within that time. If you're not willing to throw a lever and break only those eggs when the market is down, you may need to go more like 80/20.

Long-term, that will reduce your returns by 0.2-0.5%/year. Short-term, when the market drops 50%, your portfolio would only be down 40%.

Roothy

  • Stubble
  • **
  • Posts: 117
Re: bonds as retirement date insurance--premium too high
« Reply #9 on: June 12, 2018, 10:17:57 AM »
All of this is very helpful; thank you everyone.  I know this is a mental block more than anything else.  At a minimum, I'm going to move to 80/20 immediately.  But now I have another question, which I'm going to post in a new thread because it is too far off the meat of this one.  Specifically, I'm wondering what assets "count" for purposes of calculaing your asset allocation.

acroy

  • Handlebar Stache
  • *****
  • Posts: 1709
  • Age: 41
  • Location: Dallas TX
    • SWAMI
Re: bonds as retirement date insurance--premium too high
« Reply #10 on: June 12, 2018, 10:37:25 AM »
I recommend taking a look at your MINIMUM expenses that you need to withdraw from your accounts after you pass an "age of no return to work". By minimum, I mean to feed yourself, heat your home and pay for rent/mortgage/taxes.

e.g., age 65,you decide you need $10k/yr to top up your SS payments.

Then, you start a bond ladder to generate $10k/yr starting at age 65.  Yes, it is low return, but it is guaranteed, like buying your own pension, and you could sell bond in a severe need (might lose money).

^^ this. This is really the only reason to buy bonds. You are buying an annuity, a pension. Buying bonds hoping the NAV will go one way or another is pure speculation.

good luck!!

Financial.Velociraptor

  • Handlebar Stache
  • *****
  • Posts: 1275
  • Age: 46
  • Location: Houston TX
  • Devour your prey raptors!
    • Financial Velociraptor
Re: bonds as retirement date insurance--premium too high
« Reply #11 on: June 12, 2018, 02:08:32 PM »
I have discovered I have a low risk tolerance in FIRE.  When I was still working I wasn't happy unless I was 100% equity.  Now that I'm certain I never want to have to go back, I keep enough in bonds, fixed/variable rate debt, preferreds, MLPs/REITs/BDCs to cover 112% of my annual budget (before emergency belt tightenting.)  I will essentially NEVER be forced to sell equity "low".  My 'number' at death will be lower because of the low equity allocation and I am fine with that.  It will grow for a long time and still be an impressive number if I keep good health.  I'm increasingly thinking the income investments are "hold forever" and the equity portion is left to grow on its own as an organic version of the rising equity glidepath.

YMMV.

Bicycle_B

  • Handlebar Stache
  • *****
  • Posts: 1867
  • Mustachian-ish in Live Music Capital of the World
Re: bonds as retirement date insurance--premium too high
« Reply #12 on: June 15, 2018, 06:28:04 PM »
@velociraptor, @PizzaSteve, I feel you!

Being in FIRE and not wanting to NEED a job, one of the main reasons for me having some cash and bonds is to have enough to cover expenses without selling stocks, long enough not to worry much about selling stock. I specifically identify which accounts will drawn on in the 2 or 3 upcoming years, and make sure they have enough cash and bonds to satisfy me that those years feel "safe" even if there is a big (30% to 50%) bear market.