Kitces: / Managing Sequence Of Return Risk With Bucket Strategies Vs A Total Return

[macav933]

A great read again by Kitces
Managing Sequence Of Return Risk With Bucket Strategies Vs A Total Return Rebalancing Approach
Your thoughts?

Managing Sequence Of Return Risk With Bucket Strategies Vs A Total Return Rebalancing Approach | Kitces.com

 

[LOL!]

One more nail in the Buckets Strategy coffin. Actually, no more nails are needed.

 

[audreyh1]

Kitces is the name - threw me for a loop there for a moment.

 

[W2R]

Kitces is the name - threw me for a loop there for a moment.

Fixed it, thanks Audrey. Hope you don't mind that I fixed the typo in the title, macav933.

 

[samclem]

Thanks much. An "informal" bucket strategy has had its hooks in me for awhile--the idea of resisting the sale of depressed equity shares until they recover etc. This piece by Kitces does an excellent job of showing why it doesn't matter if you have buckets or not, provided that you plan to rebalance. Then he shows why it >is< so important to rebalance during the drawdown phase.

 

[donheff]

Thanks much. An "informal" bucket strategy has had its hooks in me for awhile--the idea of resisting the sale of depressed equity shares until they recover etc. This piece by Kitces does an excellent job of showing why it doesn't matter if you have buckets or not, provided that you plan to rebalance. Then he shows why it >is< so important to rebalance during the drawdown phase.

+1 Very easy to understand. It does make me think a more formal re-balancing strategy might make sense. Or psst Wellington.

 

[audreyh1]

rebalancing alone already has an astonishingly powerful effect to help avoid unfavorable liquidations, as the process systematically ensures that the investments that are up (the most) are sold, and the ones that are down (the most) are actually bought instead!

In general, this is true and I totally agree that rules against selling what is down are irrelevant when you rebalance. You don't have to worry about drawing down equities when they are down as by definition you will be buying more of them when they are down.

But in practice I ran up against another issue - when equities tank so much as they did in 2008, how much do you let your fixed income draw down to rebalance?

I have an AA and do rebalancing, in fact you could almost call me a die-hard about it. But I also have limit rule where if I don't let my fixed income portion drop under 10 years of after-tax expenses.

That helped me buy lot of equities in 2008 and early 2009 instead of standing like a deer in the headlights unable to buy. I could have bought a little more (maybe a couple of % more) to match my prior AA, but I don't know if I would have been able actually do it without the psychological comfort of many years expenses in cash and bonds.

We could truly have faced a very long multi-year period of stocks recovering, instead of the fast recovering in 2009. In that scenario I would have had the fixed income to draw on. We really did not know what we were facing at the beginning of 2009.

I'm sticking to that rule, and I hope I never face another year when I hit that limit.

My AA (and rebalancing) approach with the limit rule has been crafted to meet my psychological needs when dealing with the anxiety rebalancing sometimes incurs. So far I have rebalanced when my rules said to do so, even though at times it has been very, very difficult to do.

 

[audreyh1]

I really like Kitces' articles. He seems to be so much more "grounded" that many of the other article writers in this field. I think it's because he really checks his assumptions, and the assumptions of his peers.

 

[mrfeh]

This topic has been beaten to death over at bogleheads, and I do agree with the paper; however...

There is an assumption that the retiree has access to his/her entire portfolio. For an early retiree, this may not be the case. Furthermore, if one follows conventional advice regarding asset location, taxable accounts (which are the only option for early retirees, ignoring the 72t option) will hold equities.

This is the situation I find myself in. To have some insurance against running taxable accounts down to zero before I can access tax-advantaged accounts, I will have 3-4 years of expenses in a short term bucket, at least until I get closer to 59.5.

 

[Animorph]

Again kind of strange decision rules, best gains over the past year, ignoring the longer past history. If stocks are down 50% one year but manage a portfolio leading 2% the following year those simple rules have you selling stock for the coming year even though they have heavy losses still.

And of course if you take from whatever asset class you want, the rebalancing will completely revise that, as the paper explains. Decision rules are only useful if you let your AA vary with them. I haven't seen anyone proposing decision rules with simultaneous rebalancing.

 

[samclem]

Again kind of strange decision rules, best gains over the past year, ignoring the longer past history. If stocks are down 50% one year but manage a portfolio leading 2% the following year those simple rules have you selling stock for the coming year even though they have heavy losses still.

So, what criteria would be proper for determining if equities were still in "loss" territory? Below their purchase price ("hey, I don't track basis in my IRA!"). All time high? Priced below its 5 year average?

Regardless, as Kitces points out, if we intend to rebalance anyway (thus buying more stocks at lower price as they decline), then it really doesn't matter from which "pot" we first take the withdrawal.

 

[audreyh1]

Again kind of strange decision rules, best gains over the past year, ignoring the longer past history. If stocks are down 50% one year but manage a portfolio leading 2% the following year those simple rules have you selling stock for the coming year even though they have heavy losses still.

And of course if you take from whatever asset class you want, the rebalancing will completely revise that, as the paper explains. Decision rules are only useful if you let your AA vary with them. I haven't seen anyone proposing decision rules with simultaneous rebalancing.

This is not a problem, because by adding to equities when they were down, you reap more benefit of any subsequent gain. If they rise high enough (or bonds drop enough) to rebalance the other way then you are ahead. It doesn't matter if they haven't recovered to the prior level. It's all relative to the AA.

When I rebalanced in late 2009, it was because the stock market had rebounded such that my equity allocation became high enough to be out of balance. It didn't matter that the markets hadn't recovered their highs of 2007. I just needed to trim some of the amount I had bought when equities were much lower.

 

[RetireBy90]

I wanted to agree with Audreyh1 - I find that I put some rules in place to keep my psyche in balance also. Rules are great and they give me a framework. I have an AA and rebalance every year. I think it is important to realize what you can live with, and craft your plan to work along those limits. For example, I have a CD ladder that is not part of my AA or my retirement computations, but it is for retirement spending in case it is needed.

Well put Audreyh1

 

[LOL!]

Again kind of strange decision rules, best gains over the past year, ignoring the longer past history. If stocks are down 50% one year but manage a portfolio leading 2% the following year those simple rules have you selling stock for the coming year even though they have heavy losses still.

If stocks are only up 2% the next year, you will not have any major rebalancing move to do, so I don't really see your point.

 

[aim-high]

In general, this is true and I totally agree that rules against selling what is down are irrelevant when you rebalance. You don't have to worry about drawing down equities when they are down as by definition you will be buying more of them when they are down.

But in practice I ran up against another issue - when equities tank so much as they did in 2008, how much do you let your fixed income draw down to rebalance?

I have an AA and do rebalancing, in fact you could almost call me a die-hard about it. But I also have limit rule where if I don't let my fixed income portion drop under 10 years of after-tax expenses.

That helped me buy lot of equities in 2008 and early 2009 instead of standing like a deer in the headlights unable to buy. I could have bought a little more (maybe a couple of % more) to match my prior AA, but I don't know if I would have been able actually do it without the psychological comfort of many years expenses in cash and bonds.

We could truly have faced a very long multi-year period of stocks recovering, instead of the fast recovering in 2009. In that scenario I would have had the fixed income to draw on. We really did not know what we were facing at the beginning of 2009.

I'm sticking to that rule, and I hope I never face another year when I hit that limit.

My AA (and rebalancing) approach with the limit rule has been crafted to meet my psychological needs when dealing with the anxiety rebalancing sometimes incurs. So far I have rebalanced when my rules said to do so, even though at times it has been very, very difficult to do.

Could you unpack that some Audrey? Are you basically saying something like the?

Your annual expenses are x. You maintain a 50/50 allocation, however you never let your fixed income portion fall under 10x?

 

[audreyh1]

Could you unpack that some Audrey? Are you basically saying something like the?

Your annual expenses are x. You maintain a 50/50 allocation, however you never let your fixed income portion fall under 10x?

Yep.

In early 2009 that meant I rebalanced to 55/45, rather than the 57/43 I had been maintaining before the crash. Otherwise I would have violated the 10x rule.

 

[GTFan]

This topic has been beaten to death over at bogleheads, and I do agree with the paper; however...

There is an assumption that the retiree has access to his/her entire portfolio. For an early retiree, this may not be the case. Furthermore, if one follows conventional advice regarding asset location, taxable accounts (which are the only option for early retirees, ignoring the 72t option) will hold equities.

This is the situation I find myself in. To have some insurance against running taxable accounts down to zero before I can access tax-advantaged accounts, I will have 3-4 years of expenses in a short term bucket, at least until I get closer to 59.5.

Agree this is a great article by Kitces and clarifies a lot of the rebalancing vs. buckets question for me.

I am in the same boat as you, retiring in a month at age 51 with approx. 50/50 taxable/tax-deferred. And like you say, almost 100% equities in taxable with 3-4 years in cash. The way I look at it is not to divide the port into those two categories for rebalancing, I think that you do it holistically. Yes, you have to draw from equities that could be down a good bit in your taxable but you rebalance with the 401k to compensate, so it would end up being the same result. You just have to ensure that your taxable equity accounts have enough to cover you for 8 years (in my case) as you said. Or do like we're both doing and keep a few years' worth of cash if we run into another 50% market drop. Or do 72t distros.

I don't see it as a big issue, in other words, since with 72t's you can really start drawing in your mid-50's (or even earlier) from the 401k/IRAs if needed.

 

[ejman]

This really is a great article. It certainly confirms my direct experience with rebalancing since ER 12 years ago. I really think it's the "magic bullet" for those who depend entirely or largely on their next egg. I will consider Audreyh1's 10X fixed income tweak but what kind of fixed income?

 

[audreyh1]

This really is a great article. It certainly confirms my direct experience with rebalancing since ER 12 years ago. I really think it's the "magic bullet" for those who depend entirely or largely on their next egg. I will consider Audreyh1's 10X fixed income tweak but what kind of fixed income?

Any kind of fixed income - cash plus bonds.

 

[Animorph]

So, what criteria would be proper for determining if equities were still in "loss" territory? Below their purchase price ("hey, I don't track basis in my IRA!"). All time high? Priced below its 5 year average?

Regardless, as Kitces points out, if we intend to rebalance anyway (thus buying more stocks at lower price as they decline), then it really doesn't matter from which "pot" we first take the withdrawal.

I use my retirement portfolio projections to determine when to raise cash. I'm nominally 100% equities. If my portfolio value is at my projected 1/1/2017 value I'll take out cash for 2017 expenses, even if it's only 2015. I use percent below market peak, average of S&P 500 and International, to determine if I want to add some of that cash back into the portfolio. Ideally, I have no cash and just make monthly withdrawals for expenses, but it's been a while since it's been like that.

Yes, rebalancing neutralizes the decision rules.

If stocks are only up 2% the next year, you will not have any major rebalancing move to do, so I don't really see your point.

That was a decision rule criticism, not directed at rebalancing. Kitces set up some weak decision rules and then then made them superfluous anyway. His decision rule would have resulted in taking the full withdrawal from stocks in that case, regardless of the fact that they were still way down. Moot if you always rebalance (why have decision rules in that case?), but not good if you don't.

If I was creating similar decision rules I would probably set them to over-balance, such that the AA would tilt more heavily towards stocks when the market was far off its past peak. Certainly I have seen people on this board say they had enough bonds or cash to last through several years of a bad market without selling stocks. That would make more sense than what Kitces described, definitely change the AA temporarily, and would avoid rebalancing temporarily.

 

[Gotadimple]

Hmmm. . .Animorph,

I would say the decision rules are about distributions. How much can you take depending on the state of the market at distribution time. HOWEVER, you still need to rebalance, and that will/could generate the distribution income you need.

In practice, I determine what I need. Then rebalance to create the income.

Rita

 

[Rickt]

Hmmm. . .Animorph,

I would say the decision rules are about distributions. How much can you take depending on the state of the market at distribution time. HOWEVER, you still need to rebalance, and that will/could generate the distribution income you need.

In practice, I determine what I need. Then rebalance to create the income.

Rita

+1
Not ER'd yet, but that's what I would do