How to analyze pension SBO ? (spousal benefit option)

[Delawaredave5]

Trying to help out a friend: couple, both about 60, both will have Megacorp pensions about the same size each shortly.

There's an SBO (spouse benefit option) choice: does each of them take a $2,000 annual pension reduction to increase the spousal benefit option for the surviving spouse by $8,000.

Since both pensions are around equal and because they have other assets, the decision isn't "mitigating risk", rather "is the pension reduction a good investment for the surviving spouse ?"

My question is, how do you analyze this ?

I guess you'd do some Monte Carlo simulation with some distribution of longevity for each spouse and compute NPV ? Or compare to some life insurance product that would pay annual payments for life to surviving spouse ?

Anybody do this ? Any websites ? Thanks in advance.

 

[Big_Hitter]

The packet (or SPD) should contain information regarding the actuarial equivalence basis used to calculate the J&S options.

Generally the lower the interest rate used the bigger the haircut.

 

[Delawaredave5]

Thanks. What does "bigger the haircut" mean ?

The SPD does say, "If you choose the post-retirement survivor option, the reduction in your monthly pension is actuarially determined when you retire and start to receive pension payments. The amount of the reduction depends on the percentage of your pension you elect to have paid to your beneficiary, the Plan’s investment-return rate, and the age of you and your beneficiary.

It doesn't give a specific investment return rate, but it does show an example where it says "let's assume the investment return rate is 5%".

If true, then if the person could get greater than 5% return investing, then they'd be better off not reducing, right ? Is that a pre tax or post tax rate ?

 

[travelover]

Could you compare the reduction in pension payout to the price of a life insurance policy that would buy an equivalent single premium immediate annuity?

 

[Big_Hitter]

Could you compare the reduction in pension payout to the price of a life insurance policy that would buy an equivalent single premium immediate annuity?

you are actually buying a "reversionary" annuity - not sure quotes for those are available online

insurance is cheaper the higher the interest rate so the reduction to the normal form of payment to be actuarially equivalent to the normal form plus the reversionary annuity will be lower ceteris paribus


http://www.estimatepension.com/Annuity-Calculator.aspx

 

[Animorph]

So that is a $4k (2x$2K) reduction for as long as they both live for an $8k increase for as long as one and only one lives? Sounds better to invest the $4k unless one of them expects to check out really early. At a minimum (0% growth, 0% inflation) the survivor would have to live another 50% longer (Say 20 years as a couple and another 10 years single) just to break even. But you can put it in a spreadsheet and play with the ages.

If there is no COLA that makes it much worse, since $8k just won't be worth $8k in 20 years.

 

[Big_Hitter]

there probably isn't a COLA


it would be easier to analyze if the OP gave us monthly or annual annuity amounts to compare

 

[Delawaredave5]

So that is a $4k (2x$2K) reduction for as long as they both live for an $8k increase for as long as one and only one lives?

Above correct except the reduction would drop to $2k when one dies. There is no COLA.

Found this very interesting page: https://www.kitces.com/blog/rigorous-analysis-of-pension-options-done-right/

I used the "Pension Payout Probabilty Analysis", link below to download, and loaded the two scenarios in -- and the NPV comparing the two was essentially equal (differing by 1%, over lifetime, not annually)

The spreadsheet has actuarial data. I'm assuming/hoping the calculations are correct.

Thanks for any other look-sees.

https://www.kitces.com/wp-content/uploads/2013/09/Pension-Payout-Probability-Analysis.xls

 

[Danmar]

I had to make a similar decision. But in my case my pension was much greater than spouse so it boiled down to risk mitigation for her. Pension reduced by about 8% to go from 66.7% to spouse on my death to 100% on my death. She is 7 years younger than I so it seemed like a no brainer. Probably in the big scheme of things she would have plenty of income either way, but if just seemed fairer to allow her to get the same pension after my death, assuming I go first. If she goes first, I made a mistake.

 

[Big_Hitter]

I'd update the mortality rates to a more recent table - that 2004 period table is a killer


You would need to update columns AH, AI, AK and AL I think.

 

[Delawaredave5]

I had to make a similar decision. But in my case my pension was much greater than spouse so it boiled down to risk mitigation for her. Pension reduced by about 8% to go from 66.7% to spouse on my death to 100% on my death. She is 7 years younger than I so it seemed like a no brainer. Probably in the big scheme of things she would have plenty of income either way, but if just seemed fairer to allow her to get the same pension after my death, assuming I go first. If she goes first, I made a mistake.

I agree with you. I used the calculator -entered two scenarios: 1. $100k pension for 60 year old male, 53 year old spouse, reduced to $66k, and 2. $92k pension with 100% survivor benefit.

Spreadsheet attached. I'm surprised the NPVs of both options are so close.

Now I'm suspicious of the spreadsheet.....

Even if the NPV's are close - there's also the risk - measured by the distribution of outcomes - and I'm guessing the higher spousal benefit has less risk ?

 

[hnzw_rui]

I agree with you. I used the calculator -entered two scenarios: 1. $100k pension for 60 year old male, 53 year old spouse, reduced to $66k, and 2. $92k pension with 100% survivor benefit.

Spreadsheet attached. I'm surprised the NPVs of both options are so close.

Now I'm suspicious of the spreadsheet.....

That's to be expected because those are supposed to be actuarially equivalent. Betting the difference is just due to the mortality tables used.

Edit: NVM, didn't realize there was an insurance component to Option 2.

 

[Delawaredave5]

That's to be expected because those are supposed to be actuarially equivalent. Betting the difference is just due to the mortality tables used.

Good point. Maybe that does "prove" the model.

Question: even if actuarially equivalent, does one have more risk ?

For some reason, in this example the greater reduction/higher spousal benefit option seems like it would have more uniform cash flow probability and therefore be less risky ?

But maybe that's not true and the risk is simply "built in" the mortality table.

Thanks for everyone's help!!

 

[hnzw_rui]

Good point. Maybe that does "prove" the model.

Question: even if actuarially equivalent, does one have more risk ?

For some reason, in this example the greater reduction/higher spousal benefit option seems like it would have more uniform cash flow probability and therefore be less risky ?

But maybe that's not true and the risk is simply "built in" the mortality table.

I reckon you're probably just trading one risk for another.

How financially savvy is the couple? Who handles the investments? Would either spouse have better peace of mind and prefer just receiving a monthly check rather than managing investments? We're talking a difference in pension of around $200-400 a month. Rather than simple monetary analysis, in this case, I'd probably base the decision primarily on the human factor.

 

[Danmar]

I agree with you. I used the calculator -entered two scenarios: 1. $100k pension for 60 year old male, 53 year old spouse, reduced to $66k, and 2. $92k pension with 100% survivor benefit.

Spreadsheet attached. I'm surprised the NPVs of both options are so close.

Now I'm suspicious of the spreadsheet.....

Even if the NPV's are close - there's also the risk - measured by the distribution of outcomes - and I'm guessing the higher spousal benefit has less risk ?

Wow. That's neat. Thanks for doing that. I was 62 when I decided as that's when the pension started. Also, I would think the discount rate would be a little lower than 5%. The decision was more relationship based than financial though. If you know what I mean.

 

[kcowan]

You have to bear in mind that the 100% payout is based on the actuarial difference between the spouses. My first wife was older and when combined with my shorter projected life, the cost to ensure the difference was negligible.