The Money Mustache Community
Learning, Sharing, and Teaching => Investor Alley => Topic started by: MooseOutFront on October 10, 2014, 10:44:48 AM

My wife is likely to quit soon to stay at home with kids. Her state "pension" is vested, the cash out balance earns 2.0%, and the annuity payment starts at age 60 but is NOT inflation adjusted. There have been some cola increases by the legislature, but only 1 in last ten yrs for example.
So...
Current lump sum: $37,320. (compounds at 2.0%, can pull any time in future, roll to IRA)
Age 35
Retirement age 60 (300 months after quitting)
Monthly annuity at 60 = $1,055 (annual = $12,660)
Just on the surface using 4% SWR, it would take a lump sum of $316,500 to generate the income this annuity is offering on day 1 at age 60. Much better than an estimated 5% real return. However, with the non cola adjustment I'm having a hard time wrapping my head around apples to apples numbers.
The only other consideration is that if we don't pull it out, she could always go back to work within the same system and add to the pension under the same terms as she has now. A main benefit here, other than increasing the payout, is that it could provide a health insurance option from age 60 to age 65 if she works a few more years and makes it to 10 yr vesting.
ETA: added part about the lump sum being an option at any point between now and age 60 and that it would roll to an IRA

Let's assume that inflation will be a constant 2%. In order to value the annuity now, we use 4% + 2% = 6%. The annuity would then be worth $12,660/.06 = $211,000 at age 60 (this uses the back of the napkin valuation method you seem to prefer). If we use $37,320 as the present value and $211,000 as the future value in 25 years, then the asset grew at just over 7% annually over the 25 year prewithdrawal period (that's the nominal return, the real return is 5% with aforementioned 2% inflation assumption), which is about what we assume for long term portfolio growth around here.
So the above math seems to see imply that you would be, on average, no better off in either scenario than vice versa. However, the annuity eliminates some risk. You won't be exposed to market fluctuations in the annuity, and you're guarunteed an average return. On the other hand, the lump sum offers more flexibility and control and the possibility of above average returns.
Since the math doesn't favor one avenue or the other, go with the risk/return profile that best suits you.

Tough question Moose. I'm curious what others will have to say. I get a bit confused by inflation/real returns sometimes. I'll offer these thoughts/numbers:
You have 25 years of compounding. In 25 years that 37,320 could be worth:
5% return = 126,379
6% return = 160,173
7% return = 202,552
8% return = 255,585
9% return = 321,813
Those are not adjusted for inflation. You mention a 5% real return would not reach the $316,500, so I guess you're looking at a 78% return?
Other variable worth considering:
1) Taxes. Lump sum will not net $37,320 unless it can be rolled into an IRA.
2) Life expectancy. Your wife would have to live 30 years (6090) to achieve the benefit matched by the 4% SWR. If she doesn't, what are you left with. In the lump sum scenario you or your kids would still have the principal amount invested.
3) Pension Funding. It's backed by a very strong state, but I wouldn't want to depend on a government body to fund my retirement. I would take a pension if I had one, but I would do everything I could to gain control of the funds.
I think the key here is waiting for the annuity leaves a lot up to chance, particularly your wife's life expectancy. Taking the lump sum puts it within your control.
The use of the 4% SWR in your analysis is interesting. I guess I understand why since it's an often used metric, but does it really make sense here? It provides an 80% probability that you wouldn't run out of money. But in a lot of those 80 out of 100 scenarios you end up with an incredible sum at the end of the 30 years.
Lastly, I wouldn't factor your wife's possible return to work into this equation. From prior posts you've mentioned freelance work before reaching a 4% SWR. Unless she is pretty sure she'll go back, leave that out of the mix.

Let's assume that inflation will be a constant 2%. In order to value the annuity now, we use 4% + 2% = 6%. The annuity would then be worth $12,660/.06 = $211,000 at age 60 (this uses the back of the napkin valuation method you seem to prefer). If we use $37,320 as the present value and $211,000 as the future value in 25 years, then the asset grew at just over 7% annually over the 25 year prewithdrawal period (that's the nominal return, the real return is 5% with aforementioned 2% inflation assumption), which is about what we assume for long term portfolio growth around here.
Interesting. Are you using 4% there to value the annuity based on the 4% SWR that I mentioned?

Yep.
Let's assume that inflation will be a constant 2%. In order to value the annuity now, we use 4% + 2% = 6%. The annuity would then be worth $12,660/.06 = $211,000 at age 60 (this uses the back of the napkin valuation method you seem to prefer). If we use $37,320 as the present value and $211,000 as the future value in 25 years, then the asset grew at just over 7% annually over the 25 year prewithdrawal period (that's the nominal return, the real return is 5% with aforementioned 2% inflation assumption), which is about what we assume for long term portfolio growth around here.
Interesting. Are you using 4% there to value the annuity based on the 4% SWR that I mentioned?

Other variable worth considering:
2) Life expectancy. Your wife would have to live 30 years (6090) to achieve the benefit matched by the 4% SWR. If she doesn't, what are you left with. In the lump sum scenario you or your kids would still have the principal amount invested.
The use of the 4% SWR in your analysis is interesting. I guess I understand why since it's an often used metric, but does it really make sense here? It provides an 80% probability that you wouldn't run out of money. But in a lot of those 80 out of 100 scenarios you end up with an incredible sum at the end of the 30 years.
Thanks for taking the time to consider this. You bring up good points about the holes in trying to use the 4% SWR to get to apples to apples here. Another problem with it that is overrating the value of the annuity is that any typical 4% SWR assumption allows for inflation adjusted increases in your withdrawal, whereas this pension will be paying out the same $1000 per month when she's 80 as it did when she was 60. I really like the longevity insurance it provides, but without cola's it's tough to assume it will be worth much at age 95 when needed.

Other variable worth considering:
2) Life expectancy. Your wife would have to live 30 years (6090) to achieve the benefit matched by the 4% SWR. If she doesn't, what are you left with. In the lump sum scenario you or your kids would still have the principal amount invested.
The use of the 4% SWR in your analysis is interesting. I guess I understand why since it's an often used metric, but does it really make sense here? It provides an 80% probability that you wouldn't run out of money. But in a lot of those 80 out of 100 scenarios you end up with an incredible sum at the end of the 30 years.
Thanks for taking the time to consider this. You bring up good points about the holes in trying to use the 4% SWR to get to apples to apples here. Another problem with it that is overrating the value of the annuity is that any typical 4% SWR assumption allows for inflation adjusted increases in your withdrawal, whereas this pension will be paying out the same $1000 per month when she's 80 as it did when she was 60. I really like the longevity insurance it provides, but without cola's it's tough to assume it will be worth much at age 95 when needed.
Using my same 2% inflation assumption, the annuity will be worth about 60% of what it is today at age 60, and it will be worth about 30% of what it is today at age 90.

For this one, forget about the 4% rule, what you need is the future value to get the lump to 60, then present value to see what the stream of payments is worth at the same time. Using Cheddar's 7% return row you would have $202k in actual dollars at 60 if you took the lump and invested it. At the same point in time (2039), the net present value of the income stream (assuming continued 7% returns and your wife lives until 90) would be $157k in actual dollars. So from that point of view the lump would be a better deal, 32% better. Although most folks don't have the stomach or the means to stay 100% invested in stocks in their 70's80's and would appreciate some fixed income. This is probably as good, if not better, as any any other fixed income option.

Yes, yes, that sounds right. I need to pull out my finance class spreadsheets because NPV of income streams is just one calculation that I never committed to memory like I should have. Though I do remember it being easy to calculate if the income stream was constant, like in this case.
On the surface it seems this "pension" is close enough to being a decent value in 25 years that I don't need to be in a hurry to get it out.

Yes, yes, that sounds right. I need to pull out my finance class spreadsheets because NPV of income streams is just one calculation that I never committed to memory like I should have. Though I do remember it being easy to calculate if the income stream was constant, like in this case.
On the surface it seems this "pension" is close enough to being a decent value in 25 years that I don't need to be in a hurry to get it out.
If the lums sum only grows at 2% a year, any benefit of pulling it out will most likely decrease as time passes.

Yes, yes, that sounds right. I need to pull out my finance class spreadsheets because NPV of income streams is just one calculation that I never committed to memory like I should have. Though I do remember it being easy to calculate if the income stream was constant, like in this case.
On the surface it seems this "pension" is close enough to being a decent value in 25 years that I don't need to be in a hurry to get it out.
The problem with trying to do an exact valuation on a pension is that life expectancy is variable . As Cheddar Stacker pointed out, your wife may not even live to 60. You can combat that by using a life table to discount each year's payment by the chance of living to each age (given that she has already lived to age 30), but the round number calculations that we've been doing in this thread seem to imply that neither option is a clear winner. I think you should, instead look at the less tangible differences between your options. Will the lump sum allow you to retire significantly sooner? Are you more comfortable with the risks of the pension (pension management, inflation risk, shorter than average lifespan) or the risks of investing (sequence of returns, personal risk tolerance, longevity)?