Increase stocks with age

[Animorph]

If I understand how you modeled it, that calculation appears to have a flaw, in that your cash allocation ignores inflation for ten years. You can't count on getting an inflation adjusted $20k per year for ten years from $200,000 of fixed assets. I ran that scenario in FIRECalc, with every combo of fixed assets that they provide, and success rate was around 50% for ten years.

I'm assuming you figured the $20k/$200k was used the first ten years, then entered $40k/$800K for the next 20 years? What AA? Or did you do something else?

-ERD50

Default AA (75% equities I think). Yep, I ignored inflation for the cash supplied yearly amounts. Though I also ignored any interest earned as well. I figured that was close to a wash, though I have no idea how close the additional successful scenario was with or without the scheme.

I entered $20k withdrawals, $800k portfolio start, 30 years, -20k SS (negative to add an inflated expense) to give $40k withdrawals starting in 2023. I assumed $200k in cash would cover $20k/year (inflated) for the first 10 years. Might need low inflation, higher interest rates, or some bonds to make that happen.

 

[ERD50]

Default AA (75% equities I think). Yep, I ignored inflation for the cash supplied yearly amounts. Though I also ignored any interest earned as well. I figured that was close to a wash, ...

But it really does make a difference. When I entered a 10 year span and $200,000 portfolio into FIRECalc and tried the various fixed options, the best I got was $14,937 per year for 100% success, versus your assumed $20,000.

Even for these short time spans, a 50-50 AA did slightly better at $15,185, and 25/75 was near optimal AA - but only a bit better though at $15,192.

It doesn't seem like a conservative AA at the start really accomplishes what we would hope.

-ERD50

 

[rmark]

The high equity - low equity - rising equity pattern is what I expect, with our current high equity portfolio being overwhelmed at retirement by the value of SS and my wifes fixed pension. As the pension value falls over retirement the high equity portfolio would rise as a percentage of our assets.

 

[Animorph]

But it really does make a difference. When I entered a 10 year span and $200,000 portfolio into FIRECalc and tried the various fixed options, the best I got was $14,937 per year for 100% success, versus your assumed $20,000.

Even for these short time spans, a 50-50 AA did slightly better at $15,185, and 25/75 was near optimal AA - but only a bit better though at $15,192.

It doesn't seem like a conservative AA at the start really accomplishes what we would hope.

-ERD50

Yeah, for a really high inflation scenario $200k in cash might not give you $20k per year for 10 years. Seems like the equity side of the portfolio might be doing better than expected in that case. It would be nice to have a bit more flexibility in modeling this.

 

[ERD50]

Yeah, for a really high inflation scenario $200k in cash might not give you $20k per year for 10 years. ....

And that is the problem. If we could choose between a stock drop and rip-roaring inflation early in retirement, we could choose a defensive AA. But we can't, and FIRECALC is all about showing what happens in those relatively few bad scenarios in history. So from those runs, it's clear that 100% fixed does not let you hide from bad scenarios.

I did a few more runs, to see what happens in ten years across various AAs and a more conventional 3.5% WR (min/avg/max in $thousands):

0/100   - 372/  877/1,563 (min/avg/max)
25/75   - 434/1,022/1,817
50/50   - 427/1,176/2,262
75/25   - 413/1,338/2,869
100/100 - 360/1,510/3,766

So 25/75, 50/50 and 75/25 provide more protection (higher mins) over ten years than 0/100.

-ERD50

 

[karluk]

From the abstract

Start the retirement period with a low equity allocation, increase the allocation by 1% per year. Over 30 years go from 40% (or less) to 60% (or more).

It makes sense conceptually to lower the impact of catastrophic equity loss in the early years when it has the greatest portfolio destructive effect, and increase it in the later years when the remaining lifespan is shorter. Will people act this way? That is to say, as we age are we going to assume fewer remaining years, hence greater capability to take on equity risk? I've not seen any older retired people think or act like this. Even as they age they see their final years in the distant future and don't make financial or portfolio changes that would reflect a much shorter lifespan.

This comment is the closest to my personal reaction to the article, namely that it sounds great in theory, but there is no chance at all that the average retiree will act this way in real life. My guess is that it would work just fine in good times, such as we've seen starting in early 2009. A gradually increasing stock allocation would just cause retirees to leave more of their money in stocks as part of their annual rebalancing. Stocks have roughly doubled since 2009, so a 40% allocation in 2009 rising to a 44% allocation in 2013 would most likely have resulted in a retiree who was following this plan being a net seller of stocks over this period, just not as large of a net seller as someone who maintained a 40% stock allocation throughout the period.

The trouble comes in bad times. After a 50% market decline, rebalancing would require massive stock purchases just to get back to last year's target allocation. But now the target allocation has increased by 1%, so the retiree needs to buy even more stocks to get to the new target. I just don't see it happening.

I suspect that there are a number of people on this board who can suppress their emotions and make the required stock purchases when everyone else is selling. If you are one of them, this plan may be for you. I just don't see it ever being a common approach to portfolio management in retirement, since it requires people to react in ways that history has shown over and over again they're unlikely to do.

 

[Animorph]

And that is the problem. If we could choose between a stock drop and rip-roaring inflation early in retirement, we could choose a defensive AA. But we can't, and FIRECALC is all about showing what happens in those relatively few bad scenarios in history. So from those runs, it's clear that 100% fixed does not let you hide from bad scenarios.

I did a few more runs, to see what happens in ten years across various AAs and a more conventional 3.5% WR (min/avg/max in $thousands):

0/100   - 372/  877/1,563 (min/avg/max)
25/75   - 434/1,022/1,817
50/50   - 427/1,176/2,262
75/25   - 413/1,338/2,869
100/100 - 360/1,510/3,766

So 25/75, 50/50 and 75/25 provide more protection (higher mins) over ten years than 0/100.

-ERD50

I certainly don't advocate 100% fixed. In fact, I'm on the 100% equity side. I just carry some extra cash in the first few years. Back when I was first researching retirement there was a study that recommended a stock/bond mix and then spending down all the bonds before selling any equities. That's my thinking here. Highest bond percent for a few years just after retirement.

I don't think a 10 year FIRECalc period will show what we want. Mainly, it is still rebalancing and selling some equities and we can't tie the end results to the start of a new 20 year period.

One of the things Pfau and Kitces were saying in the latest article was a worst case scenario was typically a bear market at the start of retirement. Raising equities with age helped in this case by having more equities at the bottom of a bear market. (Sounds like historical data instead of MC in that case.)

"Reducing Retirement Risk with a Rising Equity Glide-Path"

Reducing Retirement Risk with a Rising Equity Glide-Path by Wade D. Pfau, Michael E. Kitces :: SSRN

"This study explores the issue of what is an appropriate default equity glide-path for client portfolios during the retirement phase of the life cycle. We find, surprisingly, that rising equity glide-paths in retirement – where the portfolio starts out conservative and becomes more aggressive through the retirement time horizon – have the potential to actually reduce both the probability of failure and the magnitude of failure for client portfolios. This result may appear counter-intuitive from the traditional perspective, which is that equity exposure should decrease throughout retirement as the retiree’s time horizon (and life expectancy) shrinks and mortality looms. Yet the conclusion is actually entirely logical when viewed from the perspective of what scenarios cause a client’s retirement to “fail” in the first place. In scenarios that threaten retirement sustainability – e.g., an extended period of poor returns in the first half of retirement – a declining equity exposure over time will lead the retiree to have the least in stocks if/when the good returns finally show up in the second half of retirement (assuming the entire retirement period does not experience continuing poor returns). With a rising equity glide-path, the retiree is less exposed to losses when most vulnerable in early retirement and the equity exposure is greater by the time subsequent good returns finally show up. In turn, this helps to sustain greater retirement income over the entire time period. Conversely, using a rising equity glide-path in scenarios where equity returns are good early on, the retiree is so far ahead that their subsequent asset allocation choices do not impact the chances to achieve the original retirement goal."

 

[ERD50]

I certainly don't advocate 100% fixed. In fact, I'm on the 100% equity side. I just carry some extra cash in the first few years. Back when I was first researching retirement there was a study that recommended a stock/bond mix and then spending down all the bonds before selling any equities. That's my thinking here. Highest bond percent for a few years just after retirement.

I don't think a 10 year FIRECalc period will show what we want. Mainly, it is still rebalancing and selling some equities and we can't tie the end results to the start of a new 20 year period.

One of the things Pfau and Kitces were saying in the latest article was a worst case scenario was typically a bear market at the start of retirement. Raising equities with age helped in this case by having more equities at the bottom of a bear market. (Sounds like historical data instead of MC in that case.) ...

I'm just trying to investigate this, I really don't have a solid view point either way right now.

Yes, maybe 10 years was too long, and that may give more support to somewhat higher equity AAs. I think I'll try that again a bit later with a 5 year frame.

But I also wonder, if we are talking ER we are typically looking at planning for 35-40-45 year retirements. So having a higher cash/fixed early on in a 40 year plan means you probably give up some growth in those early years. So what happens if that big drop comes in year 6 , and your portfolio has a lower buying power than it did at the start of retirement, and has to survive 35 more years?

Since FIRECalc doesn't model variable AAs, it's a little hard to say if that might be 'double counting' tragedies a bit. IOW, the 'big drop' doesn't occur after a poor period, but after a peak. But being higher in cash during that time might warp that a bit - it might have been a poor time for cash, but a good time for equities?

-ERD50

 

[haha]

I predict that this discussion will end the first time stocks fall out of bed, which may be very far away. Only thing between equities and the abyss is Señor Ben. And he seems tired lately.

Ha

 

[Onward]

Michael Kitces' latest on this subject:

Should Equity Exposure Decrease In Retirement, Or Is A Rising Equity Glidepath Actually Better? | Kitces.com

 

[bo_knows]

I certainly don't advocate 100% fixed. In fact, I'm on the 100% equity side. I just carry some extra cash in the first few years. Back when I was first researching retirement there was a study that recommended a stock/bond mix and then spending down all the bonds before selling any equities. That's my thinking here. Highest bond percent for a few years just after retirement.

I don't think a 10 year FIRECalc period will show what we want. Mainly, it is still rebalancing and selling some equities and we can't tie the end results to the start of a new 20 year period.

One of the things Pfau and Kitces were saying in the latest article was a worst case scenario was typically a bear market at the start of retirement. Raising equities with age helped in this case by having more equities at the bottom of a bear market. (Sounds like historical data instead of MC in that case.)

As far as I can tell from historical data, Starting out with say a 80/20 Stock/Cash AA, and moving to a 100/0 AA in the first 5 years, does very little to the success rate of a portfolio.

Same goes for Pfau and Kitces's method of going from 20-40% Equities and moving to 60-80% over the course of the entire retirement. Neither of these strategies seems like a panacea, but they do increase success slightly over a flat AA.

 

[ERD50]

I predict that this discussion will end the first time stocks fall out of bed, which may be very far away. Only thing between equities and the abyss is Señor Ben. And he seems tired lately.

Ha

I'm not so sure that cynicism is warranted in this case. I think the general trend of this conversation is about going with a more conservative AA early on. I don't think that is in-line with anything approaching irrational exuberance fueled by a recently rising market.

-ERD50

 

[haha]

I'm not so sure that cynicism is warranted in this case. I think the general trend of this conversation is about going with a more conservative AA early on. I don't think that is in-line with anything approaching irrational exuberance fueled by a recently rising market.

-ERD50

I must not have been clear. I am not cynical, just disbelieving. I don't judge whether or not members in this discussion are exuberant or not, or whether this would be irrational if they were. I certainly would not guess whether this posited exuberance, if it exists at all, might be caused by a rising market. Kind of a highly abstract chain!

I am just saying that the idea of increasing equities with age is a rather novel and on the surface at least, aggressive idea-(imagine being the FA explaining to the grieving and newly destitute widow that he had put her husband very heavily into stocks at age 83 to get an idea of what I mean). Also, "exuberance" is not the only behavioral sign of a very high market. Another is complacency. I use these things sparingly, because I don't have a problem with accepting the validity of PE10 as a guide. I have too many stocks, I would like to lighten up but selling is expensive with our current tax structure.

Now imagine a 50% equity fall from today, either fast or slow and grinding. Would the increase equity part of this discussion be happening? I was here the last time around, and I don't think so, to judge from the mood then. (Including a number of PMs about "This is horrible, what should I do?")

So, everybody has an opinion, just as I do. In our culture we tacitly assume a dialectical and dual structure of ideas and attitudes. This is only an assumption, not a ground level fact.

Ha

 

[samclem]

I don't know if a high allocation to equities as we get very old will catch on, but I'd say the idea of a 5-7 year "bucket" of fixed/ST bonds for the new retiree might become standard advice. After that, the "best" mix of equity exposure, bonds, fixed, annuities likely depends on your time horizon and required spending level as a % of assets.

I sure wish I-Bonds had a 4% real yield . . .

 

[ERD50]

I must not have been clear. ....

OK, sorry, I guess I misunderstood.

Now imagine a 50% equity fall from today, either fast or slow and grinding. Would the increase equity part of this discussion be happening? I was here the last time around, and I don't think so, to judge from the mood then. (Including a number of PMs about "This is horrible, what should I do?")

Sure, some people will be scared out of the market by big drops, and that last one was of historic proportions. Heck, some people are scared even w/o experiencing a drop like that! For whatever reason, I was pretty calm. I remember looking at the charts and saying to myself, 'Geez, this really is of historic proportions!' (a bit more colorful language than that). But I was aware these things happen, and though there is no guarantee, they have recovered in the past. I did some rebalancing into equities on 10/14/2008, after most of the drop occurred. And now I'm up from when I first retired (though my WR is fairly modest, with DW still bringing in a small paycheck). Didn't make any cuts in spending, still spent some planned money on some remodeling.

I guess I'm saying that the kinds of dips that we pretty much should expect from time to time shouldn't affect our long term thinking. Of course, most of us will be affected by it. But whether we are affected by it or not, that doesn't change what the numbers show us has happened historically. So I still like to look at these things to see what they tell us. I think it is worth discussing. How we each decide to act is a different matter.

-ERD50

 

[steelyman]

For whatever reason, I was pretty calm. I remember looking at the charts and saying to myself, 'Geez, this really is of historic proportions!' (a bit more colorful language than that).

That's a better reaction than I had. I just sat there dumbfounded and said "WHAT? What just happened". I remember being in a meeting in Jan 09 and thinking, "well, there that goes..."

I kept plugging away with the usual monthly contributions and it seems in retrospect to be maybe one of the most pronounced times for dollar-cost averaging, at least in my investing time. Yes, there was the tech bust in 2000, but the limits for retirement contributions were pretty low.

 

[Animorph]

Here's a quick article of how returns affect success rates. MC simulation still, but the relative importance of returns should be widely applicable. I guess the big question is how to handle it.

Wade Pfau's Retirement Researcher Blog: Lifetime Sequence of Returns Risk

"For years 31-60, we switch from accumulation to distribution, and I am showing the impact of each year's return on the maximum sustainable withdrawal rate experienced by retirees. The return in year 31 is the return for the first year of retirement, and the result in this first year explains more than 14% of the final outcome for retirees. Retirees are very vulnerable to what happens just after they retire. This result holds even more so in the real world when we consider how human capital plummets at the retirement date, as it becomes increasingly difficult to return to the workforce. Sustainable withdrawal rates are disproportationately explained by what happens in the early part of retirement. The returns experienced in the last 10 years (years 51-60) have very little impact on how much one could sustainably withdraw over retirement.

​

And so with this figure, we can see a clear demonstration of the lifetime sequence of returns risk.
"

 

[Animorph]

That's a better reaction than I had. I just sat there dumbfounded and said "WHAT? What just happened". I remember being in a meeting in Jan 09 and thinking, "well, there that goes..."

I kept plugging away with the usual monthly contributions and it seems in retrospect to be maybe one of the most pronounced times for dollar-cost averaging, at least in my investing time. Yes, there was the tech bust in 2000, but the limits for retirement contributions were pretty low.

I was trying to figure out ways to get my hands on more cash to buy equities! That's why we all have different portfolios.