Understanding Guyton withdrawal method

[rayvt]

That is, what he calls the Portfolio Management Decision Rule.
#1) Following years in which an equity asset class had a positive return that produced a weighting in excess of its target allocation, the excess allocation was "sold" and the proceeds invested in cash to meet future withdrawal requirements.

#2) Portfolio withdrawals were funded each year on January 1 in the following order: (1) cash from rebalancing any overweighted equity asset classes from the prior year-end, (2) cash from rebalancing any overweighted fixed income assets from the prior year-end, (3) withdrawals from remaining cash, ....

I can't make head or tails of this.
In #1, is this "cash" the same as the cash asset class which has a 10% allocation? Or is it a savings account that is outside the portfolio? I suspect the latter---if it's the former then wouldn't the rebalancing automatically take care of this?

In #2, I assume that item 3 refers to both the 10% cash asset and the cash that has accumulated in the side account, right?

After you've done #1, you've gotten rid of the excess in the equity asset account, so there is no longer any "overweighting" to be rebalanced in #2 item 1. So what does this mean?

The only way I can understand these is to rephrase it as:
1) Sell the excess (over the target allocation) of the equity asset classes. If that is more than the withdrawal amount, then put the excess in a side savings account to use for future shortfalls. If that's not enough then proceed with the next things to draw down. When you get to the cash withdrawal, tap the side savings account as necessary.
2) Rebalance the asset classes, but ignore the side cash account.

Am I understanding it right?

 

[2B]

Welcome to the forum. I understood Guyton much better before you asked this question.

I agree his phrasing is a bit confusing but I'll give you my take on how to execute Guyton "properly" -- IMHO.

Assume you rebalance and take your annual withdrawl on January 1. On January 1 you would execute step #1 and #2 at the same time. You would take out of your portfolio the amount allowed by the Guyton withdrawl rules. You would do this by readjusting the amounts in each asset class to meet the target for whichever Guyton portfolio you were following.

As an example, assume your portfolio was $600,000 on January 1 and your target was to have a 10% cash, 40% bond and 50% stock. The stock would also have a targeted allocation between different asset classes but I won't go there. If you intend to withdrawl $30,000, you would simply rebalance your portfolio on the basis of the remaining $570,000. You'd then have $30,000 in your separate living expense account and you'd establish your 10/40/50 by moving whatever was necessary.


BTW, for those interested I've added the link to the Guyton article.


FPA Journal - Decision Rules and Maximum Initial Withdrawal Rates

 

[rayvt]

I've played with some numbers, and I still don't get it. If you rebalance annually, then it is meaningless to say that you are funding the withdrawal from a priority-ranked list of sub-accounts. The only way you can take money from, say, an equity class while not taking money from a fixed income class is to not rebalance. Yet he implied that there is a rebalancing.

I note that the spreadsheet at http://bobsfiles.home.att.net/withdrawGuyton.zip simply takes the withdrawal from the entire account balance and then rebalances.

Guyton isn't talking about doing a 65-month rebalance interval as suggested by Bengen is he? 'cause he sure didn't say so.

 

[Kougar2]

rayvt

Note the following article:

FPA Journal - Using Decision Rules to Create Retirement Withdrawal Profiles

In the article, the author states: "At that time, the asset classes are rebalanced to the target asset allocation. (Earlier work by Guyton and Klinger (2006) used the Portfolio Management Decision Rule and did not rebalance assets each year. "

If you read the article "Retire at the Pie Shop" (Retire at the Pie Shop), you will see that this author discusses the idea of not rebalancing every year.

Hope this helps. I had the same questions when I read the paper. If you find any more clarification, please post it here.

 

[walkinwood]

RayVT,
It is confusing!
I created a spreadsheet for the different withdrawal rules in the same paper, but when I tried the same with the Portfolio Management Decision Rule I started going around in circles.

I'm still trying in my spare time. Please share any clarifications you may discover.

 

[walkinwood]

I found this link:

Retirement Article Pg 4

My confusion remains since I don't understand why step 7 is taken (or where it is mentioned in Guyton's article).

 

[rayvt]

Okay, I think I’ve got Guyton’s withdrawal funding rules figured out.

* There is a side account for cash. This account does not figure in the portfolio allocations. It is initially seeded with the starting annual portfolio withdrawal. Each month, the monthly withdrawal is taken out of this account.

You never rebalance.

At the start of each year, rule 1 (PMDR) is applied to the stock asset classes, which may (or may not) result in selling some of those assets and transferring the money into the cash account. If the application of this rule results in no transfers into the cash account, then the cash account will be depleted at the end of each year. However, if such transfers do take place, then there will be a cushion in the cash account. This cushion will be drawn upon in future years where the portfolio has a negative return. Step 3 of the withdrawal ordering is misleading---what it really means is that you’ll tap any accumulated balance in the cash account before you take money (other than that year’s return) out of the fixed income & equity portfolio.

The earnings, dividends, and interest of each asset class stays in that class and gets reinvested in that class.

In good times, you’ll be removing money (the excess earnings) from the equity accounts. When a bad stretch happens, you’ll be taking money out of fixed income assets before touching the equity accounts.

Note that there is no rule that puts money back into the fixed income class. Moreover, you don’t take money out of equities (other than the annual returns) until you’ve exhausted the fixed income classes. So I guess that somewhere along the line you might have to rebalance, either fully or partially.

* Klinger said that using the PMDR does not materially affect the shape of a retirement profile, and so simple annual rebalancing is all that is necessary.

* Beecher House uses the Guyton rules slightly modified, but then rebalances—which doesn’t make sense.

* Bob’s “pie shop” has a very complicated withdrawal rule, very similar to Guyton. But the dividends & interest are drawn off as they occur and put into the cash account for use as monthly draw—thereby reducing the amount of money that is withdrawn at the start of the year. You have to estimate the amount of dividends & interest in the coming year in order to set the start of year withdrawal amount. Also, the end of year is a partial rebalance. Any excess gain still available after the withdrawal is added to any losing asset classes, to make up the loss. He calls this “replenish laggards”. When/if the stock assets drop by a grand total of 30% he starts a quarterly partial-rebalance. Figure how much you’d have to take out of fixed income assets and do 1/4th of that amount then and for the next 3 quarters.

* Bengen says that a rebalance interval of 75 months (6.25 years) gives the highest return but with higher portfolio volatility. However, his chart has a lot of spikes that don’t seem rational, so it’s probably that there is a lot of false accuracy here.

=======================================

All in all, the PMDR rules of Guyton and variants seem awfully complex. I’m concerned that might be too complex to implement accurately. As explained by Guyton, it certainly is not clear. It has an intuitive attractiveness, so maybe they can be restated in a simpler way. “Take excess money from big winners and salt it away for lean times.”

The rest of the Guyton/Klinger rules seem reasonable and easy to apply. However declining to rebalance just feels wrong and possibly dangerous. A 5-year rebalance interval has a good feel and seems easy enough to remember.

 

[haha]

The truth is, none of this makes any sense. IMO it is a complete waste to spend even 10 minutes thinking about it. It is born of data mining, and its purpose is only to somehow differentiate the "author" of the scheme from his competitors.

Ha

 

[Nords]

Okay, I think I’ve got Guyton’s withdrawal funding rules figured out.

Ray, I know you mean well, but even reading this explanation of Guyton's rules makes my eyeballs glaze over... and I used to read Reactor Plant Manuals for a living.

At some point everyone attracted to this system is going to look up from their spreadsheets, see that it's a nice sunny day outside, and ask "Why the %&^* am I working so hard?!?"

The truth is, none of this makes any sense. IMO it is a complete waste to spend even 10 minutes thinking about it. It is born of data mining, and its purpose is only to somehow differentiate the "author" of the scheme from his competitors.

Exactly. It'll get even easier after we've done it for 20-30 years, especially when we're in our 80s. And for bonus points we'll have to explain it to our spouses.

I know-- let's have Guyton test it out on his elderly relatives!

 

[chinaco]

Ray, I know you mean well, but even reading this explanation of Guyton's rules makes my eyeballs glaze over... and I used to read Reactor Plant Manuals for a living.

 

[rayvt]

Yah, no lie!!!

After I'd read these 3-4 articles, I thought to myself that this guy is just slinging around BS and waving his hands furiously. This failed my "explain it to your mother" test right out of the gate. He can't even explain it in such a manner that sophisticated people can understand it, let alone ordinary people who don't hang out on financial web sites all day long.

It's not so much that "I mean well". I typed it out for myself to see if I understood it and pretended that I was going to explain it to my wife--whose main financial/investment interest is spending excess funds from the checking account. ;-) I figured I might as well post it here for completeness and posterity.

What was the real kicker was that Klinger (Aug 2007) wrote that "using the PMDR does not materially affect the shape of a retirement profile, and so simple annual rebalancing is all that is necessary". Which had the added benefit that it is comprehensible.

 

[2B]

I don't understand why it is so complicated. The Guyton rules are a basic stock/bond allocation with a given % taken out every year adjusted for inflation. That automatically rebalances the portfolio. Guyton allows withdrawl rates above 4%+ inflation that are adjusted upwards with good portfolio returns and down if poor returns occur. They seem straight forward to me.

 

[CuppaJoe]

As an example, assume your portfolio was $600,000 on January 1 and your target was to have a 10% cash, 40% bond and 50% stock. The stock would also have a targeted allocation between different asset classes but I won't go there. If you intend to withdrawl $30,000, you would simply rebalance your portfolio on the basis of the remaining $570,000. You'd then have $30,000 in your separate living expense account and you'd establish your 10/40/50 by moving whatever was necessary.

Thanks, 2B, I’m just getting into figuring out how I will withdraw from my portfolio. This paragraph explains a couple of things I’ve been wondering about. Maybe I’ll "get" this kind of discussion after I finishing reading a few more books. The Raymond Lucia books are next.

 

[rayvt]

2b,
The thing is, there are 2 parts to the Guyton (and Guyton/Klinger) rules. The 1st part is the one that's incomprehensibly complex. That's what he calls Portfolio Management Decision Rules (PMDR). Which is itself a misnomer, since this is really the rules & ordering for withdrawals. Why didn't he call it the Withdrawal Order Rules? Sit down and [try to] write out the steps you'd take to implement the PMDR and you'll see what I mean. For example, he never explains what he means by "sold" (referring to the excess allocation. What does he mean by the scare quotes? Is it sold or not? In the same article in different publications, sometimes he implies rebalancing and sometimes not. Yet he *can't* be doing rebalancing, since an actual rebalance would negate the careful & complex rules of what order to withdraw money.

The second part is, IMHO, the meat of the matter. The Capital Preservation Rule and the Prosperity Rule make sense and as described are easy to understand and to implement. The Inflation Rule (cap at 6%) makes sense, too---but the 2006 article modifies the withdrawal rules and therefore make the cap unnecessary.

Some people have complained that the CPR would sometimes require you to omit an annual increase and lose that increase forever. Well, guess what---several times when I was working the company skipped an annual raise due to business conditions. And they never did a catch up--all the subsequent raises were based on the actual current salary without consideration of the omitted raise. Plus, of course, the Guyton rules give you a higher SWR, so your annual withdrawals are higher than the usual rules.

 

[2B]

2b,
The thing is, there are 2 parts to the Guyton (and Guyton/Klinger) rules. The 1st part is the one that's incomprehensibly complex. That's what he calls Portfolio Management Decision Rules (PMDR). Which is itself a misnomer, since this is really the rules & ordering for withdrawals. Why didn't he call it the Withdrawal Order Rules? Sit down and [try to] write out the steps you'd take to implement the PMDR and you'll see what I mean. For example, he never explains what he means by "sold" (referring to the excess allocation. What does he mean by the scare quotes? Is it sold or not? In the same article in different publications, sometimes he implies rebalancing and sometimes not. Yet he *can't* be doing rebalancing, since an actual rebalance would negate the careful & complex rules of what order to withdraw money.

The second part is, IMHO, the meat of the matter. The Capital Preservation Rule and the Prosperity Rule make sense and as described are easy to understand and to implement. The Inflation Rule (cap at 6%) makes sense, too---but the 2006 article modifies the withdrawal rules and therefore make the cap unnecessary.

Some people have complained that the CPR would sometimes require you to omit an annual increase and lose that increase forever. Well, guess what---several times when I was working the company skipped an annual raise due to business conditions. And they never did a catch up--all the subsequent raises were based on the actual current salary without consideration of the omitted raise. Plus, of course, the Guyton rules give you a higher SWR, so your annual withdrawals are higher than the usual rules.

I think you're making it way, way too complex but I can see how you'd get confused now that I've gone back and reread the article. The PMR rules would be based on your target asset allocation and how each asset class performed in the prior year. It effectively will rebalance a growing portfolio but in a down market it would deplete your fixed income portion.

To really follow the "rules" you have to follow his asset class guidelines in his Table 1. From a practical standpoint, I considered "following Guyton" to be annual rebalancing of my asset classes every year except in down stock market years where my fixed income would be raided for my living expenses. I must confess I'm not a Guyton zealot but I like his approach.

You get to increase your withdrawls with inflation when the market is rising but its capped at 6%. Hopefully, we won't see that again but you never know.

The tricky part in my opinion was maintaining the portfolio "guardrails." Someone on the forum showed that a Guyton portfolio would have had its withdrawl rate cut by 50% during certain market periods. Even though later gains brought it back to its original purchasing power, a 50% reduction is a little tough to swallow in anyone's budget.

I've always looked at Guyton as a way to have a great travel budget that can be cut if times do get tough. It's the only model I've seen that maximizes spending and adjusts based on actual market performance.

 

[2B]

Thanks, 2B, I’m just getting into figuring out how I will withdraw from my portfolio. This paragraph explains a couple of things I’ve been wondering about. Maybe I’ll "get" this kind of discussion after I finishing reading a few more books. The Raymond Lucia books are next.

If you read my post above you can see I didn't get it right in my original example. I had simplified the method but I can't see where it would make much of a difference.

Reading Lucia is good. The buckets concept is another approach but it seems to be just another way of thinking about your asset allocation. The way I read Lucia is that he depletes the buckets in order so after 14 years you are left with 100% in stocks that you would then reset the buckets for the next 14 years. It seems you'd be much safer doing the readjustment every year or so. I'd hate my "Year 14" to be 1932 or 1974.

He pushes "non-publically traded REITs" which is an asset class I'm not very excited with. They generally have lots of fees associated with them. They are not transparent to the smaller investors involved. They are illiquid. Also, Lucia's firm will sell these to you so that makes me concerned that that's his real goal and not their long term performance.

 

[CuppaJoe]

Reading Lucia is good. The buckets concept is another approach but it seems to be just another way of thinking about your asset allocation. The way I read Lucia is that he depletes the buckets in order so after 14 years you are left with 100% in stocks that you would then reset the buckets for the next 14 years. It seems you'd be much safer doing the readjustment every year or so. I'd hate my "Year 14" to be 1932 or 1974.

He pushes "non-publically traded REITs" which is an asset class I'm not very excited with. They generally have lots of fees associated with them. They are not transparent to the smaller investors involved. They are illiquid. Also, Lucia's firm will sell these to you so that makes me concerned that that's his real goal and not their long term performance.

Interesting, 2B, when I first heard of the Lucia books, I read the Amazon.com reader comments and thought it’s not for me. Then a friend who was recently pushed out of his job/retired, thought it was the way to go for him. Anyway, I recently got it into my head to read many of the books discussed on this forum. My negative attitude about REITs has been expressed elsewhere and no planner will be able to sell them to me.

I found it difficult to shift my thinking from buy and hold to asset preservation and then planing for withdrawing. I’ve always played mind games about money so that I wouldn’t know there was more there than I imagined. One method was to differentiate between today’s NW and the number it will "hold" at and so on. For so many years, I said "I’ve done the numbers and I can’t retire yet." But change is good!