Not fond of annuities

[pb4uski]

I ran your two scenarios in Firecalc with both scenarios looking to spend $8k a year for 40 years (4% WR) to age 100. SPIA for a 60 year old pays ~6% so I used a 40 year time horizon.

Scenario 1: $200k invested 50% stocks/50% bonds. Results are as follows:

FIRECalc looked at the 105 possible 40 year periods in the available data, starting with a portfolio of $200,000 and spending your specified amounts each year thereafter.

Here is how your portfolio would have fared in each of the 105 cycles. The lowest and highest portfolio balance at the end of your retirement was $-166,024 to $1,095,060, with an average at the end of $209,190. (Note: this is looking at all the possible periods; values are in terms of the dollars as of the beginning of the retirement period for each cycle.)

For our purposes, failure means the portfolio was depleted before the end of the 40 years. FIRECalc found that 28 cycles failed, for a success rate of 73.3%.

Scenario 2: $100k stock portfolio and SPIA that pays $6k a year. Results are as follows:

FIRECalc looked at the 105 possible 40 year periods in the available data, starting with a portfolio of $100,000 and spending your specified amounts each year thereafter.

Here is how your portfolio would have fared in each of the 105 cycles. The lowest and highest portfolio balance at the end of your retirement was $-377,883 to $3,265,272, with an average at the end of $559,850. (Note: this is looking at all the possible periods; values are in terms of the dollars as of the beginning of the retirement period for each cycle.)

For our purposes, failure means the portfolio was depleted before the end of the 40 years. FIRECalc found that 20 cycles failed, for a success rate of 81.0%.

What I found interesting is that the stock/SPIA combination had much more volatility than stock/bond combination... I suspect because stocks and bonds are generally uncorrelated. It was surprising that the SPIA/stock combination had a higher success rate.

 

[nun]

What I found interesting is that the stock/SPIA combination had much more volatility than stock/bond combination... I suspect because stocks and bonds are generally uncorrelated. It was surprising that the SPIA/stock combination had a higher success rate.

Pfau came up with a similar result when he compare stock and SPIA portfolios to stock and bond ones. This is why I've taken the approach to buy into a pension plan starting at 55 so that I can have a high stock percentage that i will allow to increase over time.

 

[samclem]

I ran your two scenarios in Firecalc with both scenarios looking to spend $8k a year for 40 years (4% WR) to age 100. SPIA for a 60 year old pays ~6% so I used a 40 year time horizon.

Scenario 1: $200k invested 50% stocks/50% bonds. Results are as follows:



Scenario 2: $100k stock portfolio and SPIA that pays $6k a year. Results are as follows:

Could you explain how you modeled the SPIA? Specifically--how is the SPIAs "value" accounted for in the ending figures (since the money isn't owned by the retiree, he only has claim to the remaining annual payments, which won't be many for a 100 YO recipient). At the end of 40 years, the SPIA in Scenario 2 is worth very little, (even moreso if he dies before the 40 years, the SPIA value becomes zero, barring any "certains")and the bonds in Scenario 1 should be worth quite a bit. They both started with the same amount of stocks, maybe any higher terminal value of Scenario 2 is because the stocks weren't tapped as much to pay out withdrawals due to the higher "yield" of the SPIA (compared to the bonds in Scenario 1)?

 

[hnzw_rui]

What I found interesting is that the stock/SPIA combination had much more volatility than stock/bond combination... I suspect because stocks and bonds are generally uncorrelated. It was surprising that the SPIA/stock combination had a higher success rate.

Yep. Unlike a 50/50 stock/bond split, you're not exactly rebalancing between stock/SPIA.

Interestingly enough, here are the results of SPIA + 90/10 portfolio:

FIRECalc Results

Your spending in every year after the first year will be adjusted for inflation, so the spending power is preserved.

FIRECalc looked at the 105 possible 40 year periods in the available data, starting with a portfolio of $100,000 and spending your specified amounts each year thereafter.

Here is how your portfolio would have fared in each of the 105 cycles. The lowest and highest portfolio balance at the end of your retirement was $-339,519 to $2,637,344, with an average at the end of $464,151. (Note: this is looking at all the possible periods; values are in terms of the dollars as of the beginning of the retirement period for each cycle.)

For our purposes, failure means the portfolio was depleted before the end of the 40 years. FIRECalc found that 19 cycles failed, for a success rate of 81.9%.

 

[mathjak107]

I ran your two scenarios in Firecalc with both scenarios looking to spend $8k a year for 40 years (4% WR) to age 100. SPIA for a 60 year old pays ~6% so I used a 40 year time horizon.

Scenario 1: $200k invested 50% stocks/50% bonds. Results are as follows:



Scenario 2: $100k stock portfolio and SPIA that pays $6k a year. Results are as follows:



What I found interesting is that the stock/SPIA combination had much more volatility than stock/bond combination... I suspect because stocks and bonds are generally uncorrelated. It was surprising that the SPIA/stock combination had a higher success rate.

most study's came up with similar conclusions . the higher cash flow initially of the spia allows the equity's portion to grow longer before refilling .

in all but the best outcomes the spia had a better success rate and in many cases more money left for heirs .

folks don't realize that over time they really are depleting the cash and bonds in their own portfolio to zero in effect before refilling . we may not actually let them get to zero in practice before refilling but then we have to sell equity's sooner to refill . the end result is the same .

using spia is really the reverse .

you have your own money in the beginning when you invest on your own and over time cash and bonds trend down to zero .

with the spia you start at zero and over time get your money back , eventually getting all your money back 16 years later and then going on their dime .

if you picture letting cash and bonds go to zero and then refilling using conventional investing on your own the spia equity combo never goes to zero so the baseline income always holds so less equity's have to be sold to refill .

 

[mathjak107]

the more conservative the portfolio the more the spia can help .

other study's have shown that using an spia , your own investing and permanent life insurance can provide even higher success rates .

in all but the best outcomes or early death , 67% of the time that integrated strategy beat by term and invest the rest 67% of the time .

in 100% of the time it provided higher initial incomes and 67% of the time higher incomes and a higher balance for heirs .
in all cases buy term and invest the difference had a higher balance at the start of retirement but that is where it ended .

unlike a joint spia which is taxable , a single spia and life insurance for the spouse pays a higher draw and is 100% tax free .

that made a big difference most of the time .

 

[OrcasIslandBound]

It seems like a 50% stocks and 50% non cola spia will do better in the beginning of retirement when withdrawals from an otherwise poor market condition non-spia portfolio might be at risk of early depletion. However, later in retirement, because the spia is non-cola, inflation must be made up for entirely by the stock side. So the stocks have to increase in value double duty, once for themselves and once for the spia. Towards end of life the spia income becomes a much smaller portion of the total income. Like perhaps 1/8 vs 1/2 in the beginning of retirement. This does not seem like a good plan to me.

Sent from my Nexus 4 using Early Retirement Forum mobile app

 

[mathjak107]

not really double duty . in both cases the cash and bonds have a combined real return of zero to a negative return so equity's are needed in both cases to inflation adjust .

also a 6% annuity cash flow compared to a 4% initial cash flow on your own also already has quite a bit of inflation adjusting built in . the cash flow off the bat is already 50% higher

you can actually keep the same 4% draw and reinvest the extra 2% back in to equity's with the spia

 

[big-papa]

What I found interesting is that the stock/SPIA combination had much more volatility than stock/bond combination... I suspect because stocks and bonds are generally uncorrelated. It was surprising that the SPIA/stock combination had a higher success rate.

Some of that might be because FireCalc uses the equivalent of bond mutual funds for the bond portion - in that case you get the more immediate anti-correlation (during extreme events - over the long term, they're uncorrelated). If, however, you own actual bonds and hold them to maturity then I suspect the difference in success rates might be a bit less.

 

[mathjak107]

over time you still need to liquidate those bonds to replenish cash . so if holding until maturity in effect you are the bond fund . .

a total bond fund is more like what you would have to hold , about 25% short term , 25% long term and the rest in the middle .

 

[HadEnuff]

one element of risk that would bother me is the idea of forking over all of that money to one company, that could go belly up along the way. That is about as undiversified, as you can get.

 

[mathjak107]

even 2008 did not see one annuity or life insurance policy holder lose a penny .

even aig's annuity and life insurance business was rock solid and had nothing to do with aig's credit default swap business that needed a bail out .
they are totally isolated company's.

most states require healthy insurers to take over the client base of a failed insurer .

states also offer guarantees up to a limit .

if we have mass failings of insurers who depend on dead body's as much as financial instrument returns then i would say your portfolio will have a whole lot more to worry about .

if i decided to do them i would ladder spia's from a few company's .

 

[big-papa]

over time you still need to liquidate those bonds to replenish cash . so if holding until maturity in effect you are the bond fund . .

a total bond fund is more like what you would have to hold , about 25% short term , 25% long term and the rest in the middle .

The differences is bond funds buy/sell all of the time which makes them subject to potential annual losses. Assuming something like treasury bonds which can't default and you hold to maturity, you can't experience a loss. Not sure why other than convenience, but there are actually some bond funds that do that now..

 

[mathjak107]

i think the fact that if they are buying and selling their yields are improving over just sitting to maturity .

in your case to refill you would need to hold 1 year , 2 year 3 year ,etc ladders because if you don't you would have also sell before maturity to create cash flow .

that mix may have a lower yield then the fund manipulating their portfolio off setting any losses .

in the end i think it would be to close to matter just because you on your own need to ladder every year . don't forget your interest rates stay fixed forever on what you hold ,. as rates rise the funds yield rises .

 

[hnzw_rui]

In CA, life insurance and annuities from the are insured up to $250K for single and $320K per couple ($160K each). At 6% cashflow rate, that's good for $1.25-1.6K/month.

If I wasn't expecting pension, I would do the same as mathjak and build an annuity ladder (from different companies of course).

 

[HadEnuff]

How would an annuity ladder be built. How would it work?

 

[mathjak107]

well i would start around 70 . i would take the amount of money allocated and put a piece in each year .

THE OLDER YOU ARE AND THE HIGHER RATES THE MORE YOU GET .

that is an important thing to know

 

[nun]

well i would start around 70 . i would take the amount of money allocated and put a piece in each year .

THE OLDER YOU ARE AND THE HIGHER RATES THE MORE YOU GET .

that is an important thing to know

You could do a similar thing at a younger age if you think general interest rates are going to be rising.

 

[mathjak107]

it does not pay to start to young because rates are to low and the longer you wait the more the guaranteed payment . yeah you could collect a few more checks but in the end like social security waiting pays if you live . since buying an an annuity is a bet you will live i would go with waiting .

 

[lawman3966]

The people (well, advertisers) who attack annuities without regard to the many different annuity types and the widely varying fees that get charged seem to always be pitching some alternative product, instead of sincerely warning investors away from high fees in general.

A quick search for Ric Edelman (company is EMAP) fees yields a fee schedule within the Google search result showing that a high six figure nest egg would be subject to an annual AUM fee of about 1.8% (the fees for higher level increments to the invested amount are lower than for the initial amount of $200K). This is particularly stunning in view of EMAP's general guidance to invest in low expense-ratio ETFs.

This prompts the question as to whether it's really better to shift from high-fee annuities to high-fee stock/bond fund investments. Well, it's better for the company offering this approach, but likely no better for the consumer.

A properly backed, and suitably priced, immediate annuity strikes me as being a perfectly valid retirement financing vehicle, so long as investors bear the risks in mind and elect a suitable allocation accordingly. Annuities can be used to smooth out the volatility present in stock/bond portfolios while providing income that lasts exactly as long as the retiree does.

Lastly, I believe that the above statement that COLA adjusted annuities are not available is simply untrue. They are less common than they used to be, but can still be found. I have used the Vanguard platform several times to price inflation adjusted annuities, though in the last few months.

Whether the inflation rider is wortwhile or not is a tough one. It depends on what expectations for future inflation are. If you expect low inflation, then current rates for COLA adjust annuities won't appear worthwhile. The current TIPS rates could be criticized for the same reason. However, people still buy them, because they guarantee something that is otherwise uncertain.

The security of the insurance companies (and of the underlying guaranty associations) that offer immediate annuities is a legitimate issue. However, I don't think that vendors of competitive financial products are a reliable source of information on this subject, as they have a vested interest in steering people towards alternative products that have their own fee structures and their own risk levels.

 

[pb4uski]

Could you explain how you modeled the SPIA? Specifically--how is the SPIAs "value" accounted for in the ending figures (since the money isn't owned by the retiree, he only has claim to the remaining annual payments, which won't be many for a 100 YO recipient). At the end of 40 years, the SPIA in Scenario 2 is worth very little, (even moreso if he dies before the 40 years, the SPIA value becomes zero, barring any "certains")and the bonds in Scenario 1 should be worth quite a bit. They both started with the same amount of stocks, maybe any higher terminal value of Scenario 2 is because the stocks weren't tapped as much to pay out withdrawals due to the higher "yield" of the SPIA (compared to the bonds in Scenario 1)?

The SPIA value is not in the ending figures... the SPIA benefits offsets the annual withdrawal for $6k a year (fixed benefit while expenses grow with inflation so the initial $2k withdrawal grows with inflation on expenses).

Note that the terminal values include all retirement assets (both bonds and stocks) so they are comparable.

Yes, I suspect as you suggest that part of the reason for the higher terminal values is that the first year withdrawal for stocks is only 2% so the equities grow with compounding and mitigate stress for bad sequence of return scenarios.

Given that my fixed benefit pension is about 20% of our spending and our SS is ~50% of our spending I'm not inclined to buy a SPIA but I may consider it at some point. We have enough belt tightening in our spending that I could close a lot of the 30% gap if I wanted to.

 

[mathjak107]

we are pretty close , ss will be about 40% if we wait until 70 , small pension about 15%

 

[HadEnuff]

we are pretty close , ss will be about 40% if we wait until 70 , small pension about 15%

mathjak, only because you make reference to your waiting until you are 70, may I ask how old you are now, and how long you have been retired?

I apologize if this in an inappropriate question to ask on the forum.

 

[nun]

it does not pay to start to young because rates are to low and the longer you wait the more the guaranteed payment . yeah you could collect a few more checks but in the end like social security waiting pays if you live . since buying an an annuity is a bet you will live i would go with waiting .

The age to annuitize is a personal thing and will reflect the attitude towards risk and the makeup of your income stream. Whether it pays or not is very tied up with perception and many unknowns.......but you might well say that an early annuity that allows you to reinvest dividends from your stock mutual funds over a longer time period might work out better than waiting in some scenarios.

 

[hnzw_rui]

The age to annuitize is a personal thing and will reflect the attitude towards risk and the makeup of your income stream. Whether it pays or not is very tied up with perception and many unknowns.......but you might well say that an early annuity that allows you to reinvest dividends from your stock mutual funds over a longer time period might work out better than waiting in some scenarios.

+1. Also, if using tax deferred to buy SPIA to fund the early years, you also reduce your RMDs (unless equities do really, really, really well in which case, you've got a positive problem).

Granted, going stock/SPIA is likely similar to going stock/bond with rising equity glide path.