SS Claiming

Because I am 58 and because I am not part of a married couple and because I think SS is going to either break or be means-tested in the future I have not thought much or counted much on SS as part of my plan (I heavily discount a PV calc). But I know it's there and will have some impact. Dirk Cotton had a good article on SS claiming recently, a topic over which much ink has been spilt. In that piece, without getting too analytical, he comes up with several reasons why someone might claim early such as:

• you have a bunch of money so SS is more or less irrelevant
• you think SS is going to go bankrupt 
• you are a lower earning spouse 
• your longevity expectations are lower because of poor health
• you need the $ right away 

Most articles I have seen make a strong case for delaying SS benefits and in the end Dirk's article is no different: "All things considered, I still believe that most retirees should delay claiming Social Security benefits for as long as they are comfortable doing so. Delaying claiming is simply transferring some income from early in retirement until late in retirement for the benefit of those of us who do enjoy a very long life. It is the cheapest longevity insurance you can buy."

In my opinion, other than SS bankruptcy expectations, the biggest consideration in making the decision is longevity.  Me? I always assume a long planning period so delaying to 70 for the higher benefit amount always made sense. I mean if longevity were infinite then delay would be the only choice (not sure how we'd pay for that though). But today, after reading Dirk's article, I thought I'd look at the interaction of longevity and SS vis-a-vis the delay decision.  I'll keep the exact SS benefit values to myself for privacy but this is how I tried to do it. I'd have to ask an actuary to see if I am thinking about this the right way.  I might've simulated all this but wasn't up for the coding:

1. I probability weighted the PV of SS payments starting at: a) FRA and b) age 70 using the SOA individual annuity mortality table.[1]

2. I probability weighted the PV of SS starting at FRA and age 70 using the SS 2013 life table.

3. I probability weighted the PV of SS starting at FRA and age 70 using a binary: 100% age 58-95, 0% age 95-infinity.

I guess I was a little surprised by the result (and I should probably play around with inflation and discounting to be consistent with how I do this on my real balance sheet).  First note that #1 is conservative in that the median age at death is a little higher than #2 due to the adverse selection effects of people that buy annuities. # 2 has an earlier median age.  I don't remember exactly what the median is but the average I remember is something like 81 or 82.  #3 is very conservative and assumes a high probability of a long life.  Here is which strategy "won" in "PV x probability" terms for the assumptions I used:

1. More or less dead (no pun intended) even,

2. Collect at FRA. The NPV diff was about 10k in favor of FRA,

3. Delay to age 70. The NPV diff was about 25k in favor of age 70.

So what does that mean in my plan? If SS was all I had, I would have to take it early or at FRA.  If I had enough assets to cover 67-70 I'd probably wait but based on the NPV results it probably doesn't matter unless I have really long life expectation.  Since the NPV results as a % of net worth are not too big of a deal either way I'll probably get a check-up at 66 and make a call then. My default assumption is to delay.  Right now I keep a placeholder on an "actuarial" balance sheet equal to the NPV of a stream of SS payments inflated at 2% and discounted at, I think,  3 1/2 or 4% and then I divide that by 2 because I don't trust the program or the politicians all that much.


---------------------------------------------

[1] inflated at 3%, discounted at 5%. Probability is the probability of still being alive at a given age.