Understating Sequence of Return Risk

[pb4uski]

I just thought of a good reason to use an income approach to withdrawals, ie only spend divs. Sequence of return risk is zero if you assume divs don't get reduced. Even if you relax this assumption I would guess the chances of a material div reduction is less than a material market decline. Also, div paying stock is on average less volatile.
So if you are worried about sequence of return risk but don't want to increase your FI allocation, this approach might reduce your risk while keeping the upside?

Only drawback is that your WR is low so you may never save enough to be able to retire, but other than that...... good plan.

 

[mathjak107]

trying to spend only dividends or only interest is really not my idea of a retirement income .

the issue is dying with to much money unspent may not be the greatest plan either . ideally you want to enjoy your money and spend some of it down while still leaving legacy money for heirs .

this is what a good retirement withdrawal plan does .

what about inflation adjusting ? dividends may not track your personal rate of inflation which can differ vastly from the cpi ?

you still have sequence risk . the value of your dollars compounding after each dividend payment will be less at the gate each quarter unless you are offsetting the decrease in share price from the dividend being paid out by a enough of a gain . in an extended down turn this can be an issue .

in reality getting a 6% total return with a 2% dividend and 4% appreciation is no different than drawing that 2% against a portfolio that has 6% appreciation and no dividend . they both are effected by the same thing .

 

[pb4uski]

While it has worked out well for me (retired at the end of 2011 and our 60/40 portfolio is up by 4-5 years of withdrawals despite recent market correction) so I think i'm out of the woods .... if I knew then what I know today I might have been more conservative (say 40/60 rather than 60/40) to mitigate sequence of returns risk and then transition back toward 60/40.

 

[nun]

trying to spend only dividends or only interest is really not my idea of a retirement income .

the issue is dying with to much money unspent may not be the greatest plan either . ideally you want to enjoy your money and spend some of it down while still leaving legacy money for heirs .

this is what a good retirement withdrawal plan does .

That is one definition of success, and the one that is used by most retirement advisors. However, some people might want to leave money to heirs or just might not see joy in spending money unnecessarily. Personally my plan does not involve spending dividends, capital gains distributions or interest; they are all reinvested and I hope to leave a large inheritance to my grand nieces and a a bunch to some local charities.

 

[nun]

While it has worked out well for me (retired at the end of 2011 and our 60/40 portfolio is up by 4-5 years of withdrawals despite recent market correction) so I think i'm out of the woods .... if I knew then what I know today I might have been more conservative (say 40/60 rather than 60/40) to mitigate sequence of returns risk and then transition back toward 60/40.

A friend once said to me that auxilliary verbs have no place when investing. So never say you "should have" done something or "would have if........" there lies madness.

 

[daylatedollarshort]

trying to spend only dividends or only interest is really not my idea of a retirement income .

the issue is dying with to much money unspent may not be the greatest plan either . ideally you want to enjoy your money and spend some of it down while still leaving legacy money for heirs .

This somewhat gets into the idea does spending more money really equal more happiness? I think for us the happiness studies do hold true and spending more money would not make us significantly happier. We aren't exactly eating cat food now. I know many posters here have multiple retirement income streams and can live well off one or two. I don't need to spend like Bill Gates to have a pleasant retirement.

I like the idea of never having to worry about money while I'm alive and having money leftover to leave to our favorite charities after we are gone. Personally, I would rather see our money go to an elephant sanctuary than a Ferrari dealer.

 

[mathjak107]

if you got 20k in dividends and they were in your retirement account and you reinvested them and took 20k from cash or bonds instead is that any different ? of course not .

in the end no matter how you generate that income stream it does not matter . trying not to touch principal just means you are taking a low withdrawal rate against your total portfolio but in reality how you comprise that income stream it is the same .

as far as the dividend aristocrats go .

Today's aristocrats are a very different list than what they were back in 1990 . why do you think that is ?

out of the original 26 only 7 are left today .

ten were removed because they either cut or suspended their dividend , so much for that belief , four were removed for reasons that were not listed and the remainder were taken over at some point. So at least ten of the 26 ended up not producing that dividend as expected yearly even though they did for decades . .

basing a retirement income on only that belief would be a poor plan .. of course i would buy them today , but no i would not develop an income stream i had to live on solely off of them . i would have lots of stocks , bonds and possibly reits .

i would just keep a low withdrawal rate if i wanted to preserve principal .

by the way , just drawing the proverbial 4% inflation adjusted off a 60/40 mix has left you with principal untouched 30 years later 90% of every rolling time frame since 1926 and more than 2x what you started with 67% of the time .

 

[nun]

if you got 20k in dividends and they were in your retirement account and you reinvested them and took 20k from cash or bonds instead is that any different ? of course not . in the end no matter how you generate that income stream it does not matter . .

Taking the dividends from a retirement account means they are taxed as income. From a non retirement account they are treated as dividends

 

[daylatedollarshort]

in the end no matter how you generate that income stream it does not matter .

I agree with that. I think the different philosophies are some retirees feel that if they don't spend as much as they can then they won't be as happy as they could have been, while others are okay continuing to save money in retirement, whether it is reinvesting interest or dividends or saving money in some other way.

 

[Danmar]

Only drawback is that your WR is low so you may never save enough to be able to retire, but other than that...... good plan.

Doesn't need to be though. My equity portfolio is yielding about 4%( was lower before this correction). Over several years probably averaged 3.5-3.75% which might be a reasonable SWR. Real problem is the FI portion of a balanced portfolio which yield is pathetically low now. This doesn't affect me though, because I treat my pension as a FI proxy and don't have FI otherwise.

Surprised more people who are concerned about sequence of return risk don't consider an income approach.

 

[Birchwood]

From my understanding "Sequence risk" affects mainly the first 20 years after we retire. The whole market is going into a cycle, up and down, how long or short it takes affects our withdrawal pattern. Thus if we have a long recession during the early part of retirement, it can be a tough miserable few years.
What it practically means is that we should withdraw less when the market is down and probably get more when it is up to make a balance.

 

[Danmar]

trying to spend only dividends or only interest is really not my idea of a retirement income .

the issue is dying with to much money unspent may not be the greatest plan either . ideally you want to enjoy your money and spend some of it down while still leaving legacy money for heirs .

this is what a good retirement withdrawal plan does .

what about inflation adjusting ? dividends may not track your personal rate of inflation which can differ vastly from the cpi ?

you still have sequence risk . the value of your dollars compounding after each dividend payment will be less at the gate each quarter unless you are offsetting the decrease in share price from the dividend being paid out by a enough of a gain . in an extended down turn this can be an issue .

in reality getting a 6% total return with a 2% dividend and 4% appreciation is no different than drawing that 2% against a portfolio that has 6% appreciation and no dividend . they both are effected by the same thing .

Agree mostly, except we are talking about sequence of return risk. If you are forced to sell something to sustain your lifestyle at a bad time, this is sequence of return risk.

If I can live on the divs(I can) and the divs are close to a SWR (they are) I don't see the sequence risk? Agree I have div reduction risk and my portfolio may not be as diversified as many would like. Maybe these risks are comparable in size to sequence risk but I don't think so.

Agree that I will probably need to do something to prevent leaving too high a legacy for my daughter. But I can deal with this by liquidating some principal at an advantageous time. Not be forced into doing so.

Actual results for 18 years of investing. Total CAGR for total return about 12% about 3-3.5% of this divs rest(8.5-9%) cap appreciation.

 

[mathjak107]

Taking the dividends from a retirement account means they are taxed as income. From a non retirement account they are treated as dividends

actually michael kitces looked in to this and what he found was if equity's throw off dividends the tax deferred account was a better choice then the lower tax brackets of a taxable account for 2 reasons .

the fact that the dividends can compound tax deferred over time beat the tax difference from the special capital gains rates and the compounding capital gains tax deferred in the retirement account beat the equity's in the taxable account .

the 2nd reason is turnover . the more the turnover in equity's the worse they will do in a taxable account and that is true of funsds too .

you can see in your head after decades of paying taxes in the taxable account on dividends the lesser amount loses more compounding ability than the tax difference .

https://www.kitces.com/blog/asset-l...e-account-versus-ira-depends-on-time-horizon/

 

[nun]

actually michael kitces looked in to this and what he found was if equity's throw off dividends the tax deferred account was a better choice then the lower tax brackets of a taxable account for 2 reasons .

the fact that the dividends can compound tax deferred over time beat the tax difference from the special capital gains rates and the compounding capital gains tax deferred in the retirement account beat the equity's in the taxable account .

the 2nd reason is turnover . the more the turnover in equity's the worse they will do in a taxable account and that is true of funsds too .

you can see in your head after decades of paying taxes in the taxable account on dividends the lesser amount loses more compounding ability than the tax difference .

https://www.kitces.com/blog/asset-l...e-account-versus-ira-depends-on-time-horizon/

I was thinking of the withdrawal phase. I assume that most people have equities in their after tax accounts as they will have filled up their bond allocation in retirement accounts. When it comes to withdrawing funds taking qualified dividends from taxable accounts allows for some tax savings strategies depending on your income level and if you reinvest dividends in a retirement account the tax deferral compounds.

 

[mathjak107]

exactly the case that kitces looked under the hood and found that really may not be the best way to do it . dividends distributions and turnover can wipe away any tax advantage having the equity's in the after tax account . having low yielding bonds and cash may be a waste in the deferred account as well . unless you were in the zero capital gains bracket the savings from capital gains rates may be negated out .

this is just the opposite of what old school used to think . but they never looked at it with distributions and turnover in investments so the tax savings always appeared a no brainer . but not so .

 

[nun]

exactly the case that kitces looked under the hood and found that really may not be the best way to do it . dividends distributions and turnover can wipe away any tax advantage having the equity's in the after tax account . having low yielding bonds and cash may be a waste in the deferred account as well . unless you were in the zero capital gains bracket the savings from capital gains rates may be negated out .

this is just the opposite of what old school used to think . but they never looked at it with distributions and turnover in investments so the tax savings always appeared a no brainer . but not so .

Given the choice between the immediate tax savings of 0% tax on qualified dividends from a taxable account and income tax on dividends from retirement accounts I think I'll take that rather than a generalized analysis.

 

[hnzw_rui]

exactly the case that kitces looked under the hood and found that really may not be the best way to do it . dividends distributions and turnover can wipe away any tax advantage having the equity's in the after tax account . having low yielding bonds and cash may be a waste in the deferred account as well . unless you were in the zero capital gains bracket the savings from capital gains rates may be negated out .

this is just the opposite of what old school used to think. but they never looked at it with distributions and turnover in investments so the tax savings always appeared a no brainer . but not so .

I think growth rate was probably the biggest determining factor for "efficient asset location". Bond and other fixed income in taxable accounts could certainly be quite inefficient at the 6-10% interest rates of decades past. In today's low-rate environment, not so much.

Of course, the Kitces model assumes one is in the 25% marginal tax bracket and doesn't take into account the potential tax torpedo when you start drawing SS/pension nor the effects of RMDs. To be honest, I think tax drag mostly matters if you're re-investing dividends and capital gains distributions. If you're using the distributions (and maybe even selling shares) for your annual spending, then I think the tax treatment of LTCG and q-divs along with tax loss harvesting might be preferable. There's also the small matter of the complexity of drawing from tax-deferred accounts penalty free if you retired, say, in your 40s or something.

Really, if pensions and other income will put you in 25% or higher marginal tax bracket, best place to have stocks and other high growth vehicles is the Roth.

 

[nun]

Really, if pensions and other income will put you in 25% or higher marginal tax bracket, best place to have stocks and other high growth vehicles is the Roth.

Yes that's perfect for lowering tax.

There are some practical income considerations when you are spanning the gap to 59.5 and beyond even if you do a 72t. One of those is keeping immediate tax to a minimum so that you maximize the money you have to spend. So if I can take qualified dividends from a taxable account any pay 0% tax (assuming I keep my income in the right tax bracket) that sounds a lot better than some potential deferred benefit.

In the real world most people have equities in tax deferred accounts along with fixed income and I would expect them to also have some equities and cash in taxable accounts.....certainly if they are going into ER.

 

[pb4uski]

Given the choice between the immediate tax savings of 0% tax on qualified dividends from a taxable account and income tax on dividends from retirement accounts I think I'll take that rather than a generalized analysis.

+1

 

[papadad111]

From my understanding "Sequence risk" affects mainly the first 20 years after we retire. .

This is mostly dependent on the age one retires.

So I ask- For true early retires, those in 40's for example, it would seem that the risk of inflation is far greater over the course of a 40-45 year retirement than series of return risk.

Would it make sense that , in the above context, holding a 60/40 AA (45 years of retirement to cover) with the 40 percent earning next to nothing is a lot riskier *in total* than staying at 100 percent equity asset allocation yielding an approximate 2 percent dividend yield. ?

We've just been through 8 years of near zero interest rates.

 

[mathjak107]

inflation is part of the sequence of return risk equation when spending down . it is already built in to the result . in fact the worst case scenario we had to date 1965/1966 did not fail based on just market returns . it was inflation that did it in .

but just by definition sequence risk when spending down has to include inflation adjusting .

1966 stocks 10.23 - -bonds 7.85 -- rebalanced portfolio 9.56- - inflation 5.38

for comparison the 140 year average's were:

stocks 8.39--bonds 2.85%--rebalanced portfolio 6.17% inflation 2.23%

the biggest danger though is the first 15 years . the returns above were pretty ok . buit looking at the 15 year results you can see what made them worst case , inflation did .

1966-stocks minus .13%--bonds 1.08%--rebalanced .64%-- inflation 5.38%

 

[utrecht]

From my understanding "Sequence risk" affects mainly the first 20 years after we retire. .

That's kind of like saying you only lose a football game if you score less points than your opponent.

"Sequence risk" implies there is more risk at the beginning of retirement. Well, the first 20 years IS retirement for the majority of people. Its at least half for most everyone so obviously the first 20 years are critical. I would say once you get past the first 5 years unscathed, you're in pretty good shape

 

[Danmar]

That's kind of like saying you only lose a football game if you score less points than your opponent.

"Sequence risk" implies there is more risk at the beginning of retirement. Well, the first 20 years IS retirement for the majority of people. Its at least half for most everyone so obviously the first 20 years are critical. I would say once you get past the first 5 years unscathed, you're in pretty good shape

Yes, but it depends on how long your retirement is expected to be. I feel good that I have escaped sequence of return risk for the first 9 years of my retirement, but I am only 65 and hopefully have a long retirement left. So, I still should be concerned. Equivalent to a 65 year old retiring today.

 

[hnzw_rui]

Yes, but it depends on how long your retirement is expected to be. I feel good that I have escaped sequence of return risk for the first 9 years of my retirement, but I am only 65 and hopefully have a long retirement left. So, I still should be concerned. Equivalent to a 65 year old retiring today.

Ah, but how much was your starting retirement portfolio and how much is your portfolio worth now?

I reckon the situation is much less risky for someone who starts off with $1M and after 9 years has an inflation-adjusted portfolio of $1.5M versus someone who starts off with $1M but only has $500K inflation-adjusted after 9 years.

 

[Danmar]

Ah, but how much was your starting retirement portfolio and how much is your portfolio worth now?

I reckon the situation is much less risky for someone who starts off with $1M and after 9 years has an inflation-adjusted portfolio of $1.5M versus someone who starts off with $1M but only has $500K inflation-adjusted after 9 years.

It's a little more complicated for me. For the first 5-6 years of retirement I was still cashing out deferred compensation awards to fund our living expenses. Some of these cash outs were used to buy vacation properties. If I add these back my portfolio would be a little bigger than when I retired(perhaps by inflation). The awards were MTM in my initial retirement portfolio. This despite the fact we were spending heavily on other things like travel during this period. So all in all not a bad result especially when you consider the impact of the financial crises during the initial period of retirement.