Buckets and Bears

[twaddle]

ER's know two things about stocks:

1) Stocks have a higher return than bonds in the long run

2) Retiring into a bear market with a high allocation to stocks is a potential retirement killer

So, how do you balance these two factoids? Some have talked about buckets, some have talked about starting with a higher bond allocation and moving into a higher stock allocation over time. But we've never modeled these strategies.

Until now! bob90245 over at diehards.org has put together a nice spreadsheet that compares a bonds-first withdrawal strategy to other withdrawal strategies.

Harvesting Withdrawals in Retirement

The results are what you'd expect: withdrawing bonds first results in a high stock allocation over time, and it also allows you to make it through those early bears with fewer scars. Try setting the IWR to 4% and the retirement year to 1929, 1968, and 1969.

The worst scenario appears to be a 1966 start, but a relatively high allocation to small stocks helped.

[Moemg]

Interesting but that's based on a 50/50 split .Wondering how it would work at say 30 bonds /70 stock ?

[twaddle]

Move the slider and find out. You can alter bond/stock mix and large/small mix.

[cute fuzzy bunny]

Quote:

Originally Posted by twaddle 1) Stocks have a higher return than bonds in the long run

Better than that...

[twaddle]

Yes, bunny, we know. You should obviously have a 100% allocation to stocks in retirement then, right?

Try the simulator to find the answer.

[cute fuzzy bunny]

Oh dont be so defensive. Nothing in my post suggested any particular allocation was better or worse. Just a statement of returns after taxes and inflation.

[twaddle]

Some people make the logical leap that since stocks have higher returns, that means they should have more stocks.

They think that bonds merely dampen volatility. Obviously that helps, and FIREcalc shows that bonds will improve your SWR.

But the dispersion of returns matters too. And it matters most to ERs who are withdrawing from their nest egg in bear markets. It's interesting to consider what a high initial allocation to bonds might do for you.

[cute fuzzy bunny]

Quote:

Originally Posted by twaddle Some people make the logical leap that since stocks have higher returns, that means they should have more stocks.

I didnt say that either, but it might be a pretty smart logical leap for people without enormous portfolios that have to live off of their retirement funds for 40-50 years.

[twaddle]

Don't hate me, bunny. Please just play with the simulator and let me know what you think.

I have no vested interest here -- I just think this is the first interesting twist on FIREcalc we've seen in a while.

[cute fuzzy bunny]

Aww, the bunny loves you twaddle. And i'm a bunny, not a shark, so you can stop thrashing at the water.

I dont have any vested interest here either. I just put up a picture to help people get a frame of reference.

[Animorph]

I would normally favor 100% equities, which is what the bonds first quickly devolves to. But my own retirment calculations (just using constant returns in this case) showed very little penalty for keeping a few initial years of expenses in bonds/cash, providing it was spent first. With the state of the market and economy, and an equity balance a little larger than I needed to retire, that sounded like a good idea. Especially since I have a kid in college and another on the way in a couple of years. My early-retirement expenses are going to be huge compared to the normal base spending. Seems like a really cheap insurance against a couple of bad years of market returns right at the start.

I plan to harvest money from the equities whenever the balance exceeds the planned withdrawal point using the fixed growth model, if the market cooperates. Hopefully that will lead to a bonds/cash allocation after the market has run up a bit and allow smaller or no withdrawals when it is down. I don't plan to have a fixed bond/cash allocation. We'll see.

Dan

[haha]

Quote:

Originally Posted by twaddle I have no vested interest here -- I just think this is the first interesting twist on FIREcalc we've seen in a while.

I'll say. Starting years after the mid-sixties are not for the faint of heart.

My take home. If you retire into a bull market, don't worry, be happy. You would have been better off with 100% equities, and the most volatile equities at that, but you didn’t have a crystal ball, did you?

If you retire into a bear market or a stagnant high inflation environment, be sure to have enough fixed assets to get you into a bull market. Your fixed assets will be gone by the time you get there, but who cares, the bull market will kite your equities, especially if you have a lot of volatile small caps.

Sometimes the bad times last just too long for the cavalry (bull) to arrive, and you should have had some rentals or a job, you loser.

This tool would make me examine my whole premise if I were a total return asset allocator.

Ha

[donheff]

I don't see why the data would call for a "bonds first" strategy. WHat if you retire into a bull? Better a "bonds first if you are in a bear market." The take-away would seem to be "keep a fair amount of safe funds (good bonds, CDs, whatever) to get you through a bear whenever it arrives."

[twaddle]

Don, it sounds like you're describing the "high first" strategy which he directly compares to the bonds-first approach.

Basically, you want to leave your stocks alone for as long as you can. And you want a bond buffer to get through an early bear.

It'd be interesting to me to see how this strategy would have worked in Japan-1990. I'm not sure 50% bonds would have been enough to make it through that bear.

[haha]

Quote:

Originally Posted by donheff The take-away would seem to be "keep a fair amount of safe funds (good bonds, CDs, whatever) to get you through a bear whenever it arrives."

While that would seem to make sense, it is not what this paper found. Of course it is based on a very limited data-set. But "keep a fair amount of safe funds to get you through a bear whenever it arrives" was specifically refuted here by the relatively poor performance of rebalancing strategies.

Ha

[windsurf]

QUOTE=haha: While that would seem to make sense, it is not what this paper found. Of course it is based on a very limited data-set. But "keep a fair amount of safe funds to get you through a bear whenever it arrives" was specifically refuted here by the relatively poor performance of rebalancing strategies.

That sounds reasoned and i am impressed by your assertion. What is your present age and asset allocation?

[cute fuzzy bunny]

My take was that I oughta pay off my debt, build an asset allocation that produces a dividend payout that meets my reduced spending and also maintains pace with my experienced level of inflation.

I know thats all backwards...figure out what you need, reduce it where possible, then craft something bulletproof to provide those needs. Seems the standard is to horse around with allocations and how you withdraw them, and then cobble together a system that fixes all sorts of problems that could have been avoided in the first place and worry about the spending later on.

Absent ten years of double digit inflation or mass defaults/bankruptcies, since a dividend strategy eliminates the need to sell depreciated assets, its bulletproof to bears and takes advantage of bulls to a good degree.

Should we get ten years of double digit inflation or mass defaults/bankruptcies, I suspect pretty much everyone with any plan will be thoroughly screwed.

The two lines of thinking arent mutually exclusive though. I pretty much have two 'buckets', one that I'll spend from for the next 15-20 years and one for past that. They're pretty much mirror images of each other in my portolio...the short one creates dividends and should pace inflation. The other one has a lot of high volatility, high return components.

Someone that doesnt have $3M+ probably needs to pay more attention to the balance between "risk" (whatever that means to you) and return. Carry too much fixed income in your "short bucket" in anticipation of events we may never see and you might create your own little late life "depression"...

[free4now]

I wonder how much of this is data mining artifacts. Everyone including the paper's authors agrees that withdrawing bonds first is overall a more volatile approach overall than rebalancing, because you wind up with 100% stocks. So it should be more likely to fail in worse-than-history cases. As the study shows, it may do slightly better in no-worse-than-history cases, but who is to say the future won't be worse than history?

[haha]

Quote:

Originally Posted by free4now I wonder how much of this is data mining artifacts.

I think it is all data mining, or at least stretching to get conclusions from insufficient data. But then so is most retirement oriented finance literature. The biggest value here is in the failure examples; and I think they are quite valuable. It is really eye-popping to move the slider and watch the lines hit zero.

BTW, I think Bunny has a great plan. Just retire twice as rich as you need to be, so you can live off 1/2 your stash without selling securities.

Who says we don't have geniuses on this board?

Ha

[TickTock]

Interesting research! I hope to see more comparisons of alternate strategies against the rebalancing model. Thanks for the link, twaddle.