Should you rebalance your retirement portfolio

[nun]

Here is a study of various withdrawal strategies from a retirement portfolio

http://www.bobsfinancialwebsite.com/pdfs/Is_Rebalancing_a_Portfolio_During_Retirement_Necessary.pdf

The conclusion is that the best plan is to spend your bonds/fixed income allocations down to zero before touching your stocks. This leaves you with a 100% stock portfolio which you might go with or then rebalance into bonds if you just can't deal with the volatility.

This is presented as some amazing discovery, but basically we all know that a portfolio heavy in stocks has historically outperformed one biased towards bonds so by spending down the bonds and letting the stocks ride it's not difficult see that it might be a high return approach. So why isn't the conclusion of the study that a 100% stock portfolio is best? and why isn't risk considered? Such planning is like an engineer designing without considering the 100 year event.

I'd go with a 100% stock portfolio if I had a pension or bought a SPIA to cover my expenses, but failing that I'd still rebalance a 50/50 retirement portfolio as I believe it gives good return with significantly less risk than the spending bonds first approach.

 

[LOL!]

nun has linked an article from the Journal of Financial Planning by Spitzer & Singh. The article is not by bob who posts frequently on the Bogleheads forum.

I don't know why Spitzer & Singh really ignore risk, but they surely make a point in their conclusion that they do. OTOH, I think many retirees are way too conservative in their asset allocations.

 

[nun]

nun has linked an article from the Journal of Financial Planning by Spitzer & Singh. The article is not by bob who posts frequently on the Bogleheads forum.

I don't know why Spitzer & Singh really ignore risk, but they surely make a point in their conclusion that they do. OTOH, I think many retirees are way too conservative in their asset allocations.

This paper isn't so much about the irrelevance of rebalancing, it's really about keeping a larger fraction of a portfolio in stocks and the risk they emphasize is that of not producing enough growth in your portfolio to maintain your withdrawals. The analysis of SWR vs AA for the various withdrawal strategies is interesting. But their conclusion is really that a 100% stock portfolio is the best way to go. This may be supported by the numbers, but to be able to sleep at night I think I'll still hold bonds and rebalance so I'm prepared for that 100 year event.

 

[lark_L]

Sure 100% give you better return, but you don't have time to make up for a big loss when you're in retirement. Somewhere I read, maybe Otar (?), that these assumptions work OK for accumulation, but not for "deccumulation".

 

[nun]

Sure 100% give you better return, but you don't have time to make up for a big loss when you're in retirement. Somewhere I read, maybe Otar (?), that these assumptions work OK for accumulation, but not for "deccumulation".

The article is saying that more stocks and bond spending first work best in the decumulatiom phase. I assume that a similar analysis for accumulation would also conclude that more stocks are better and also that rebalancing isn't worthwhile.

 

[rescueme]

I think many retirees are way too conservative in their asset allocations.

1. What would you consider "too conservative"?

2. Are you a retiree? In other words, are you expressing your opinions while still having income from a j*b, or are you actually using your retirement portfolio to meet your income needs in retirement (as I do)?

- Just wondering...

 

[chinaco]

This article shows some data to support some ideas.

The PIBR analysis suggests that the Bonds First withdrawal strategy is likely to leave a larger portfolio at the end of 30 years than rebalancing would. A reasonable recommendation to a retiree would be to first harvest bonds and then harvest stocks. This strategy provides outcomes at least equivalent to rebalancing with respect to shortfalls and has the added benefit of a potentially larger estate. An advisor, however, still needs to be cognizant of the behavioral aspects associated with portfolio volatility.

The paper is essentially about a bucket style approach with varying sizes for the fixed bucket (number of years of income).


The success of the approach may depend on when one locks in the stock gains before beginning the withdrawal phase of those assets... (i.e., choose the exit time x years before they will be needed so one does not get caught with Reverse Dollar Cost Average)..... But, I did not see comments in the article about the withdrawal phase of the stock assets.... I assume they would be convert to fixed assets when the bonds are depleted. Did anyone see information about the stock exit technique?


However, there is some disagreement.

In Otar's book "unveiling the Retirement Myth" he show a 10 year dedicated asset strategy to be inferior to some other methods.

 

[Independent]

I was having trouble matching the text and the numbers. When I look at Table 1, it appears that rebalancing is very slightly better than "bonds first" for withdrawal percentages of 3% and 4%. For example, in the 50/50 portfolio, rebalancing failed 1 or 8 times, while bonds first failed 2 or 9 times.

That seems to say that rebalancing is the "safer" strategy.

Then I noticed that for withdrawal percentages of 6% and 7%, rebalancing fails more often (41 and 58 vs. 32 and 46). That may be true, but it's relevant only for people who are comfortable with withdrawal rates that fail 32% of the time. Very few of us fit that risk profile. As nun says in the OP, if you're a risk taker, you're better off in stocks.

 

[LOL!]

1. What would you consider "too conservative"?

2. Are you a retiree? In other words, are you expressing your opinions while still having income from a j*b, or are you actually using your retirement portfolio to meet your income needs in retirement (as I do)?

- Just wondering...

Fair questions.

1. Too conservative is anything less than 30% stocks.

2. I am semi-retired. I am withdrawing some from our portfolio to make ends meet. But to be fair, my part-time income is not small.

 

[rescueme]

Fair questions.

1. Too conservative is anything less than 30% stocks.

2. I am semi-retired. I am withdrawing some from our portfolio to make ends meet. But to be fair, my part-time income is not small.

Thanks for your answer.

Just as a comparison, I'm retired (a bit over four years) and me (along with DW, who is still employed - her desire, but expected to retire in the next year or so, at the same age) are at a 50/50 mix (e.g. equity/bond-cash split).

So in your definition, we're not conserative at all; at least we're down from the 90/10 AA of our employment years.

As to income (as related to me, alone), I draw 99% of my retirement income from my retirement portfolio, some of which I "transmuted" to an SPIA (10%, at the time) when I retired back in early 2007.

My target (and which I have no problem in achieving) is 100% of my pre-retirement net income, adjusted for my personal "rate of inflation"...

Yeah I/we are doing OK...

 

[nun]

I was having trouble matching the text and the numbers. When I look at Table 1, it appears that rebalancing is very slightly better than "bonds first" for withdrawal percentages of 3% and 4%. For example, in the 50/50 portfolio, rebalancing failed 1 or 8 times, while bonds first failed 2 or 9 times.

That seems to say that rebalancing is the "safer" strategy.

Then I noticed that for withdrawal percentages of 6% and 7%, rebalancing fails more often (41 and 58 vs. 32 and 46). That may be true, but it's relevant only for people who are comfortable with withdrawal rates that fail 32% of the time. Very few of us fit that risk profile. As nun says in the OP, if you're a risk taker, you're better off in stocks.

Excellent points. The more I think about the paper's results the more obvious they become, eg you need a high stock percentage to support large annual withdrawals.....duh. As I'm planning on 2% withdrawals I just don't need anything more aggressive than 50/50 and I choose that as the sweet spot on the risk reward curve and because I'm a middle of the road investor with little ability to predict the future. My income needs will be fully met by rental income, SS, UK state pension and a small company pension. The trick for me is finding the correct strategy to bridge the gap between ER and SS age. The rental income covers 50% of my income requirements, but should I buy a short duration annuity to cover the rest or use a CD ladder?

 

[Animorph]

That's sort of what I did when I retired. Except I used about 3 years of cash instead of bonds. We've all heard that the first few years of retirement are crucial if you are expecting to take a fixed income + inflation from the portfolio. If you can bridge those first few years with cash/bonds while stocks grow, 100% stocks may not look too bad.

I ended up reinvesting the cash as the market declined into 2009. I was 100% stocks at the bottom of the market, plus some added borrowed money from the HELOC. So 100%+ I guess. I was selecting mutual funds/ETF's that were slightly over allocation to sell to meet expenses about 6 months after the bottom.

However, I have also regained my cash cushion lately since my portfolio value is now above my retirement projections. I've been selling funds and raising cash. I may be spending/reinvesting that cash over the next two to three years if the portfolio doesn't exceed the recent peak for a few years.

A two bucket approach, but with a long-term 0% allocation to the cash/bond bucket. As long as there is cash available and the portfolio is not doing better than projected, I spend the cash down.

Worked great for a 2007 retirement.

 

[Animorph]

Excellent points. The more I think about the paper's results the more obvious they become, eg you need a high stock percentage to support large annual withdrawals.....duh. As I'm planning on 2% withdrawals I just don't need anything more aggressive than 50/50 and I choose that as the sweet spot on the risk reward curve and because I'm a middle of the road investor with little ability to predict the future.

The lower the withdrawal ratio, the less chance a 100% equity portfolio runs out before you want it to. I expect FIRECalc indicates an 100% equity SWR of greater than 2%. Then look at the ending portfolio values for 100% stocks and 50% stocks. Which case is "safer" when you're 100 years old? I always think of that as paying for rejuvination treatments, but whatever, more money is more options. That's my main motivation for 100% stocks. And I do have two sons to use anything I don't spend. It doesn't seem right to hand them an estate portfolio that's been 50% bonds for the previous 30+ years. They should have been in all equities when they were young.

But then, your 50% bonds seems much closer to the norm than my 0%!

 

[kcowan]

How does this study relate to the Efficient Frontier that showed be best overall returns from an 80/20 stocks/bonds portfolio for successive ten-year periods? I know they did not model withdrawals, so is this study just about that phase?

 

[growing_older]

One problem with this approach, and similar bucket strategies, is that there is a time when most assets except the highest risk (equity) have been used and you must now convert the equity or spend it directly. If this happens to be a deeply depressed time in equity markets, you must lock in a bunch of losses, more or less depending on your particular flavor of bucket strategy. If your withdrawal rate is low enough, you can probably get away with taking some distributions from an equity only portfolio, and then the higher long term expectations for equities likely work in your favor as markets recover, but if your SWR is approaching 4% this risks depleting the portfolio. I strongly prefer stock/bond balances.

 

[veremchuka]

When I read the op post I was shocked that Bob would advocate such a philosophy but I opened the link and saw he was not the author.

It struck me that this is similar to the Buckets of Money approach using 2 not 3 buckets. But in the BoM you replenish buckets 1 and 2 from 3 which is equities.

100% equities sounds like a recipe for disaster. 2008? No rebalancing possible. What do you do if the market does a 2008 and stays down for 24 or 30 months vs 18? We all know that you can draw down assets to the point of their inability to recover. I doubt anyone in retirement could sleep with a 100% equity allocation, maybe 1 in a million? 2008 and 2009 spanked me hard and I learned just how intolerant I am of losses which I thought my tolerance was pretty high, it is not!

I did not read the article because it is so against what I believe I consider it a waste of time. Not to say they are wrong or it might work for you but I assure you it does not work for me!

 

[Independent]

The trick for me is finding the correct strategy to bridge the gap between ER and SS age. The rental income covers 50% of my income requirements, but should I buy a short duration annuity to cover the rest or use a CD ladder?

We used CDs and I-bonds in the bridge, they happened to be after tax investments for us. I didn't want to take the market value risk on large withdrawals that I knew would come soon after retirement.

There was a poll on this board on this topic some time ago. IIRC, about a third said they had dedicated assets for the bridge, about a third said they were just planning higher withdrawals from their normal AA, about 10% said they hadn't thought about the issue, I've forgotten the other option.

 

[dtbach]

100% equities sounds like a recipe for disaster. 2008? No rebalancing possible. What do you do if the market does a 2008 and stays down for 24 or 30 months vs 18? We all know that you can draw down assets to the point of their inability to recover. I doubt anyone in retirement could sleep with a 100% equity allocation, maybe 1 in a million? 2008 and 2009 spanked me hard and I learned just how intolerant I am of losses which I thought my tolerance was pretty high, it is not!

I did not read the article because it is so against what I believe I consider it a waste of time. Not to say they are wrong or it might work for you but I assure you it does not work for me!

+1 This approach may work for a computer (which has no emotions) but would put most people through a wringer of worry in real life. I'm not sure who said it but it went something like "yes the market will always come back, problem is, you might die waiting for it"