Asset allocation in retirement

[Cobra9777]

I can't see Ferri advocating 30s/70b for an early retiree....

I don't think Ferri is making a case for 30/70 as a definitive final answer for all retirees. He says it should be a "starting point" for a discussion about AA for the average retiree; just as 60/40 should be the starting point for the average accumulator. His last paragraph says that adjustments up or down from those starting points is "...fully understandable and acceptable after a thorough assessment of safety needs, income, longevity and estate planning considerations in retirement."

Interestingly, by the time I was a year or two from retirement, I had shifted to 30/70 under the same rationale Ferri mentions in the article:

There is no economic reason for a person to take more investment risk than necessary once they’ve accumulated enough money for retirement. The focus should be on the minimum amount of risk needed to achieve an income required in retirement.

However, after reading extensively on the subject (including this forum), and experimenting with various inflation and longevity assumptions in my retirement spreadsheets, I concluded that, as a 52 year-old retiree, I needed considerably more equity. In addition, we have some significant unknowns about LTC, both for my in-laws near term and ourselves longer term. It's not unreasonable to conjure up a worst-case scenario (inflation, longevity, LTC) where 30/70 failed miserably.

And as I said earlier in the thread, we have a substantial portion of expenses covered by two pensions (one non-COLA) and rental income. I generally consider the real estate as part of my equity allocation, although it has characteristics of fixed income as well. Either way, counting the PV of pensions as bond-equivalents results in ~50/50 overall, excluding cash reserves. I'm very comfortable with that balance of return and volatility for this stage of ER, and considering the circumstances, I don't think Ferri would object either.

 

[audreyh1]

For those looking at allocations to stick with in retirement (not the rising glide path case), you should be aware of graphs like the first one below that show when equity allocation drops below 45% what you can safely withdraw over a long period drops too.

This was in one of Bengen's books according to the referring article.

 

[REWahoo]

^ What audrey1 said.

Here's a chart from FIRECalc illustrating the same allocation information. Go lower than 40 - 45% in equities and the 30 year success rate drops off significantly.

 

[mathjak107]

quote:
There is no economic reason for a person to take more investment risk than necessary once they’ve accumulated enough money for retirement. The focus should be on the minimum amount of risk needed to achieve an income required in retirement.
--------------------------------------------------------------------------------------

i am a big believer in that quote . many have pensions and for them the pay check never really stops.

they are generally in a different group when talking about the above quote.

they are still in their accumulation stage pretty much if all they need are some slight withdrawals or the withdrawals are really fun money.

for many they are not investing for themselves and their homemade income stream , many are really investing for heirs and not so much themselves and their income.

my goal now is not to grow poorer and i want more downside protection than upside potential.

 

[mathjak107]

^ What audrey1 said.

Here's a chart from FIRECalc illustrating the same allocation information. Go lower than 40 - 45% in equities and the 30 year success rate drops off significantly.

the updated trinity combined with monte carlo simulations had concluded anything above 35% would be pretty successful especially at 3.50%.


30 years, withdrawal 4% inflation-adjusted (no fees):
Considering all allocations, the best success probability is 87%.
That 87% success probability is achieved at any allocation that is at least 60% stocks.
An 86% success probability is achieved with stocks allocation 40% or 50%.


Reduce withdrawal to 3.5% inflation-adjusted (no fees):
Considering all allocations, the best success probability is 92%.
That 92% success probability is achieved at any allocation from 20% stocks to 80% stocks.

Reduce withdrawal to 3% inflation-adjusted (no fees):
Considering all allocations, the best success probability is 98%.
That 98% success probability is achieved only with 0% stocks.
A 97% success probability is achieved with any allocation from 20% stocks to 80% stocks.

 

[MichaelB]

I find it surprising Mr. Ferri did not mention withdrawal rate or portfolio survivability in an essay about asset allocation for retirees. Perhaps he expects returns for the equity portion will be high.

 

[mathjak107]

the origonal studies looked at 1926 to 1996. that was only 41 retirement time frames.

as of 2009 we now have 55 retirement time frames and looks at retirement dates from 1926 to 1980.


surprisingly there are still only 2 failure periods , 1965 and 1966.



the flaw is that the really bad years for the markets showed up in the later years of the retirement time frame of the more recent look.. that only has a very small effect at that point.

a newer retire reting more recently like the y2k retiree would have those bad return years early on creating very different results when their numbers are finally in.

 

[rayvt]

There is no economic reason for a person to take more investment risk than necessary once they’ve accumulated enough money for retirement. The focus should be on the minimum amount of risk needed to achieve an income required in retirement.

There are 2 big problems with this statement.

The first is that he implicitly equates volatility to risk.
It isn't. Volatility is not risk.

The second is that he completely ignores inflation risk. Well, a risk is still a risk even if you ignore it.
On average, stocks will float up with inflation. But bonds do not.
You will get the par amount when they mature, but the buying power will be lower. And if you sell them before they mature, then the value will float up and down just like stock prices go up and down -- so what have you accomplished?

 

[enjoyinglife102]

While many on this forum have had the discipline to ride out significant equity fluctations, lots of people outside of this forum have not. Mr. Ferri is just presenting that a 30/70 portfolio should produce a solid return that is less likely to result in panic selling, so it's something to consider. If one can enjoy life at those spending levels, than why not avoid possible stomach churning volatility? It's similiar to Bernstein's advice that if you've already won the game, why keep playing?

 

[audreyh1]

While many on this forum have had the discipline to ride out significant equity fluctations, lots of people outside of this forum have not. Mr. Ferri is just presenting that a 30/70 portfolio should produce a solid return that is less likely to result in panic selling, so it's something to consider. If one can enjoy life at those spending levels, than why not avoid possible stomach churning volatility? It's similiar to Bernstein's advice that if you've already won the game, why keep playing?

Ultimately the long retirement issue is inflation.

If you can amass a large enough portfolio that your withdrawal is small enough such that your portfolio survives. But this can requires a larger portfolio so it may mean working more years to reach that goal. That might be a significant trade off. Looking at the above chart for a 40 year survival you would need a 14% larger portfolio if you wanted 30% equity allocation instead of 40%.

Otherwise you're just ignoring the long-term risk to reduce the short-term volatility.

Bernsteins "won the game" approach also ignores the effects of inflation long-term. In his case, he has the "won the game" piece in cash and short-term fixed income for 25x needs. And only then can you put amounts above that in equities. Well a 100% bond portfolio only supports a 3.33% withdrawal rate for 30 years compared to a more balanced approach supporting 4.25% or better. That requires a larger portfolio to support the less volatile approach and enjoy the same standard of living - 28% larger. And the longer the survival time frame, the worse the divergence. It may mean the difference between retiring early and not.

If someone has 20 or 25 years or less to be retired, I don't see a problem with sticking with 30% equities. But early retirees need portfolios to last 35 to 40 years or more.

 

[mathjak107]

There are 2 big problems with this statement.

The first is that he implicitly equates volatility to risk.
It isn't. Volatility is not risk.

The second is that he completely ignores inflation risk. Well, a risk is still a risk even if you ignore it.
On average, stocks will float up with inflation. But bonds do not.
You will get the par amount when they mature, but the buying power will be lower. And if you sell them before they mature, then the value will float up and down just like stock prices go up and down -- so what have you accomplished?

the problem with risk vs volatility is volatility is only volatility until you run out of time. then it becomes risk.

if we have not recovered by the time you need to refill spending money then the risk of selling at a loss becomes very real.

if anyone ever told me back in 2000 that the balance on my equities portion would barely be up 1.77% a year real return on average 15 years later with dividends reinvested i would have thought they were nuts.

 

[mathjak107]

Ultimately the long retirement issue is inflation.

If you can amass a large enough portfolio that your withdrawal is small enough such that your portfolio survives. But this can requires a larger portfolio so it may mean working more years to reach that goal. That might be a significant trade off. Looking at the above chart for a 40 year survival you would need a 14% larger portfolio if you wanted 30% equity allocation instead of 40%.

Otherwise you're just ignoring the long-term risk to reduce the short-term volatility.

Bernsteins "won the game" approach also ignores the effects of inflation long-term. In his case, he has the "won the game" piece in cash and short-term fixed income for 25x needs. And only then can you put amounts above that in equities. Well a 100% bond portfolio only supports a 3.33% withdrawal rate for 30 years compared to a more balanced approach supporting 4.25% or better. That requires a larger portfolio to support the less volatile approach and enjoy the same standard of living - 28% larger. And the longer the survival time frame, the worse the divergence. It may mean the difference between retiring early and not.

If someone has 20 or 25 years or less to be retired, I don't see a problem with sticking with 30% equities. But early retirees need portfolios to last 35 to 40 years or more.

more and more studies are showing retirees need a whole lot less inflation adjusting than we think or are told.

the drop off in spending from 70 to 80 is so large that what you stop doing and buying more than offsets any inflation increases on what you continue to spend on..

we pick up again after 80- because of healthcare costs and gifting but over all inflation adjusting is no where near what we predict.

 

[samclem]

Yes, now that Mr Ferri has been advising clients for awhile, it's possible he has gotten the same "bug" that afflicted Dr Bernstein: They know what the math and the historical records say, but they have also seen how their clients actually behaved. So, perhaps they are giving advice that matches that behavior, not what would (theoretically) be best.

I'd prefer the advice to be free of embedded behavioral considerations, instead put those in the "notes, warnings, and cautions" that go along with the numbers.

A cynic would say that Dr Bernstein and Mr Ferri are just trying to minimize adverse reactions from their clients, thereby putting their own "comfort" ahead of objectivity. And, as a bonus, recommending that clients invest more conservatively means they'll have lower expected returns and may need to save more. Mr Ferri is compensated, at least in part, according to the size of his clients portfolios.

more and more studies are showing retirees need a whole lot less inflation adjusting than we think or are told.

A thousand people = a thousand scenarios. I can do a better job of estimating how I might want to live when I'm 75 than if I just use the "averages."

 

[bmcgonig]

Yes, now that Mr Ferri has been advising clients for awhile, it's possible he has gotten the same "bug" that afflicted Dr Bernstein: They know what the math and the historical records say, but they have also seen how their clients actually behaved. So, perhaps they are giving advice that matches that behavior, not what would (theoretically) be best.

I'd prefer the advice to be free of embedded behavioral considerations, instead put those in the "notes, warnings, and cautions" that go along with the numbers.

A cynic would say that Dr Bernstein and Mr Ferri are just trying to minimize adverse reactions from their clients, thereby putting their own "comfort" ahead of objectivity. And, as a bonus, recommending that clients invest more conservatively means they'll have lower expected returns and may need to save more. Mr Ferri is compensated, at least in part, according to the size of his clients portfolios.

I thought if that too, but then I thought that a higher stock/bond ratio would be in the best interest of the advisor since the portfolio should grow more so the AUM would be more with time. Also some portfolio managers do charge a different fee (less) for the bond portion than the stock portion.


Sent from my iPad using Early Retirement Forum

 

[roromarks]

56 year old early retiree from San Diego

Hi,

I'm interested in hearing feedback on the analyses shown below (see link to entire article, along with excerpt). As a 56 year old early retiree currently invested 40 stocks / 40% bonds / 20% cash/CDs (to allow breathing room for fluctuations in markets and possible investment in additional rental property), I'm wondering if going even more conservative (to 20 - 30% stock) for the first 5 years is a prudent decision. Thoughts?

Asset Allocation for Early Retirees | Hull Financial Planning

I created a market and inflation simulation to see if this couple could make it to age 95 without running out of money. I used historical market returns for stocks, ranging from an annual loss of 44.2% to an annual gain of 56.8% and a median return of 10.5%, and for corporate bonds, ranging from a minimum annual gain of 2.5% to a maximum annual gain of 15.2% with a median gain of 5.2%, to determine this couple’s returns over time. Each year, the simulation picked a random return for stocks and bonds as well as choosing a random amount inflation, ranging from 10.5% deflation to 18% inflation with a median inflation of 2.8%.
I ran the simulation 10,000 times to see how often our couple had money at age 95.

They had a positive net worth at age 95 89.4% of the time. 50% of the time, they had more than $32.5 million, and 50% of the time, they had less than that amount.
When I construct model scenarios, I usually like to aim for a 90% or higher success rate. This is right on the borderline, but if this couple was a client of mine, I’d give them the go-ahead with a few warning signs to look out for.
However, recent research by Dr. Wade Pfau, CFA shows that you’re most vulnerable to poor market performance in the first few years of retirement. His research was limited to 30 years of earning income and 30 years of retirement, but in that research, he showed that returns in the first year of retirement explained more than 14% of the final outcome for those retirees. Since the risk is mainly poor market performance in those first few years of retirement – bad years compounded with withdrawals when the total amount of money that the retirees have is expected to be at its lowest – I decided to tweak the strategy to determine if I could improve on this couple’s outcome.
Instead of using the 110 – age asset allocation strategy that would lead to the couple being invested 70% in stocks the first year of retirement, I had them only 20% invested in stocks and 80% invested in corporate bonds for the first five years of their retirement. The results?

This time, they had money left over 93.4% of the time, and their median net worth was $30.6 million. A shift lower by about $2 million to get well into the safety zone is a tradeoff that I’d be willing to make.
If 5 years is good, then 10 years should be better, right?
Yes and no.

The couple had money left over 94.6% of the time – a much smaller increase in safety. They also had a median net worth of $28.4 million. They paid more for a smaller increase in the cushion.
This makes sense intuitively, as, aside from investing completely in annuities, it’s nearly impossible to gain complete safety when investing in the market. Additionally, the longer that the couple is so heavily invested in bonds, the more likely it is that inflation could catch them.
The counterintuitive approach that Dr. Pfau advocates for the first few years of retirement for people in their sixties seems to hold true for early retirees as well: shift into more conservative investments for a few years and then shift back into more aggressive investing. There’s a balance in asset allocation in retirement between being too conservative and having your portfolio decimated by inflation and being too aggressive and having unfortunate downturns in the market just as you retire destroy the value of your portfolio. Based on my analysis, portfolio protection in the first few years of early retirement seems to be a prudent approach in improving the chances of having your money outlast you.
Is this surprising information? What do you think? If you go conservative early in retirement, will you have the fortitude to dial the risk back up in a few years? Let’s talk about it in the comments below!

 

[audreyh1]

The risk of an early retiree having a low equity allocations is that inflation will probably hurt you badly in the long run.

However, there might be a reason to start at 30% equities as long as you let your equity allocation rise over time. Check out Kitces "equity glide path" work in this area. he is trying to address the risks in the first 15 years. As Clients Age, Stock Up on Stocks: Kitces

I see you are mentioned the Pfau approach. There are already quite a few threads discussing the rising equity glide path in this site.

I would never drop below 30% at the start however. Long-term portfolio survival doesn't do will with low equity exposure.

Personally I probably won't drop below 45% equities until I'm well into my 70s. I'm at about 53% now.

 

[Animorph]

The first few years in retirement probably are critical, if you didn't retire at a 100% successful WR, but I'm not sure you can do a whole lot about it with a static or open-loop allocation. I think the ideal is to reach a 100% "safe" withdrawal rate as soon as you can, if you didn't retire at a low WR.

That could be done with 100% stocks and a great first few years without a big subsequent dip. An early market dip would be trouble. If you go super conservative for the first few years, you are less vulnerable to a portfolio dip, but you are also not making a lot of progress towards that safer WR. You are vulnerable longer. And a great first few years might not get you to that 100% safe WR. So 100% stocks and a favorable market looks good. Super conservative is good if the market dips early, but then you want to become aggressive for the recovery rather than stay conservative. I think that's where Dr. Pfau's rising equity allocation gives a small boost, even if he is doing it regardless of market conditions.

Hopefully we'll see more on simple dynamic allocations (algorithmic, not guesswork) in future retirement studies. I haven't managed to try it out at this point.

 

[Dash man]

I noticed you didn't have any down years in your bond market performance scenarios. Considering we've been in an extremely long bond bull market, interest rates are at historic lows, and government, consumer and corporate combined debt is at historic highs, I think it would be prudent to adjust your assumptions. There are many concerns about liquidity in the bond markets that could lead to a major financial crisis. Handcuffs put on the banks by Dodd-Frank may prevent actions by the banks to avoid a crisis. Just some food for thought.