NPV differences in when SS is taken

[marko]

Let's not get hung up on semantics. Needing the money was not the outcome. Using the money (as you did) (both for "fun" and not to diminish your portfolio balance) was the outcome.

"Extra cash" and cash is the same difference as the words, proactive and active ... there is none.

I think we're getting hung up on semantics.

 

[gayl]

Held off to 70 due to family longevity & pension covers 100% of expenses (daily + 4 cruises per year + annual gifts to kids). Helps having home paid off + prop 13. Eventually have to take MDR

 

[Jack Cumming]

I took my SS several years ago, so this does not affect me, however seeing this table on the internet this morning surprised me; thought the differentials would be greater. I have not verified the calculations.
View attachment 63185

 

[Jack Cumming]

This is not an actuarial determination and does not take into account the health condition of the beneficiary. If you believe you will live longer than average, wait to apply. If you think you might die sooner, take the benefits earlier, assuming only money maximization is guiding your thinking. Financial planners love the certainties implicit in hypothecating a cutoff date while ignoring that more and more people are living to 105 and beyond. It's best to prepare for the worst and hope for the best. If you take the planner's oldest death age, 95 in this table, don't imagine that the planner will be there to take responsibility when you outlive the "assumption" and enter the costliest years of old age.

 

[Acollector]

I started another thread where I was looking at what to do and came to similar numbers/conclusions. Starting mine at 63 and 9 months.

Flieger

Does your decision consider a spousal option and 50% of yours with non employed apouse AND fact her half of yours after your death?

 

[samcat]

I figured it out myself in a pretty simple spreadsheet and determined 62 for me, 70 for my husband. He is older and made more money.

I then went to opensocialsecurity.com and it agreed with me. I was so happy that my simple spreadsheet gave me the same answer.

Yes, we would get more money with me claiming later but not that much more.

 

[JBTX]

I'm not sure the 4% real is actually a good number to use. But maybe someone else will know.

Not really. Typically what you would use is the risk free real rate, which if you look at TIPS it’s over 2.0%. However using higher real rates is useful for scenario analysis, and if you truly plan to reinvest the difference from filing earlier and are willing to accept the additional risk, it is useful information.

 

[Montecfo]

Not really. Typically what you would use is the risk free real rate, which if you look at TIPS it’s over 2.0%. However using higher real rates is useful for scenario analysis, and if you truly plan to reinvest the difference from filing earlier and are willing to accept the additional risk, it is useful information.

I hear you. And classically that makes sense. But deferring is not in fact "risk free" when you are looking at a 24% cut here in a few years.

I wonder how the analysis looks if we assume Congress is unable or unwilling to act to fix this problem.

It would skew toward claiming earlier I would think.

 

[Flieger]

Does your decision consider a spousal option and 50% of yours with non employed apouse AND fact her half of yours after your death?

Spouse already started hers at 62 + 4 mos. I considered survivor benefits as well.

Flieger

 

[indiajust]

If you die on schedule, it does not matter when you file, it evens out. It is the over lived and underlived folks that need to be cautious.

And that is all other things equal. Which is not true. Inflation and systemic overpromising guarantee change is coming. Roll the dice with that.

 

[JBTX]

If you die on schedule, it does not matter when you file, it evens out. It is the over lived and underlived folks that need to be cautious.

And that is all other things equal. Which is not true. Inflation and systemic overpromising guarantee change is coming. Roll the dice with that.

While true, that is for the average American. There is good evidence that those who are in the top quartile of income / wealth live longer - by a pretty dramatic margin vs the bottom 20%. Given the average poster here is significantly more well off than the average American, all else equal, the numbers favor deferring, especially for married couples. However there are always exceptions.

 

[Perryinva]

Like the OP said.. the difference is pretty small as a cumulative total if you hit 85. And very few people are thinking about how smart they were because they optimized waiting until 70, at 85. Smaller amount longer, larger amount shorter.

DW just plain WANTED hers at 62. I compromised and went for FRA, when I realized Roth conversions were a waste for us. The amount I ended up getting was more than I expected to get waiting until 70, when I was 55, thanks to those 3 high COLA years in a row. Good enough. It’s pretty hard to find anyone that says they regret taking it when they did, unless they waited until age 70 and got a bad diagnosis a few years later. It is way too easy to think about those ages in your 50s as being normal. When you hit late 60s and 70, the reality of how old 85 is, is way more apparent.

 

[Lewis Clark]

Like the OP said.. the difference is pretty small as a cumulative total if you hit 85. And very few people are thinking about how smart they were because they optimized waiting until 70, at 85. Smaller amount longer, larger amount shorter.

DW just plain WANTED hers at 62. I compromised and went for FRA, when I realized Roth conversions were a waste for us. The amount I ended up getting was more than I expected to get waiting until 70, when I was 55, thanks to those 3 high COLA years in a row. Good enough. It’s pretty hard to find anyone that says they regret taking it when they did, unless they waited until age 70 and got a bad diagnosis a few years later. It is way too easy to think about those ages in your 50s as being normal. When you hit late 60s and 70, the reality of how old 85 is, is way more apparent.

Not always that hard.

Now that they are in their late 80s, both my mother and my father-in-law regret taking SS at 62.

 

[marko]

Not always that hard.

Now that they are in their late 80s, both my mother and my father-in-law regret taking SS at 62.

If I'm lucky enough to reach ~86 and hit my break-even, I hope that I have those regrets.

 

[gauss]

Don't forget that there are individual psychological / personality-driven effects at play here too.

If you like to take your "pain" early and delay gratification -- which I suspect may be true for a majority of long term readers of this board - this may sway your decision and/or how satisfied you are with it.

Also those demographic tables about life expectancy may not apply well when applied to our (the readers of ER.org) particular population.

As I like to say "We aren't Most People".

-gauss

 

[big-papa]

Not really. Typically what you would use is the risk free real rate, which if you look at TIPS it’s over 2.0%. However using higher real rates is useful for scenario analysis, and if you truly plan to reinvest the difference from filing earlier and are willing to accept the additional risk, it is useful information.

Yep - There might be a reason to use a different rate that isn't necessarily TIPS of a single number of years. And that is if you're withdrawing your projected SS benefits from another account before you start SS and you want your total, before-tax income in the years before SS begins to be the same after SS begins.

At that point, then you want the discount rate (in real terms) for SS to be the same as the expected returns of the account from which you're withdrawing. Depending on the source of equivalent pre-SS income, this might also be time variable.

Siamond explores this in his 2 part series on the bogleheads blog on Time Value of Money (TVM) concepts:

Early Retirement and Time Value of Money (Part 1) – Financial Page

Early Retirement and Time Value of Money (Part 2) – Financial Page

If you've already decided when you want to start SS and you decide that you want to use a TIPS ladder as the bridge, then there's no need to do any TVM math. But if you've decided that this is what you want to do AND you want to figure out the best age to start SS, then this requires more work. A 5 year TIPS ladder will usually have a lower aggregate yield than a 10 year TIPS ladder, for example. And tools like opensocialsecurity.com can only do its analysis with a single discount rate. So you probably need to run several scenarios using different discount rates which would reflect different length TIPS ladder bridges to arrive at a final answer.

There's also an opportunity cost involved in doing this as well which may or may not be acceptable to an individual investor since, one way or another, additional money is coming out of investments to fund the pre-SS equivalent income.

I didn't do the analysis like this for our case because, frankly, I didn't think of it at the time. I just used opensocialsecurity with its default discount rate, with the only option being to use the expected lifetime tables for healthier individuals. Result was my DW starting SS 1 year before FRA and me waiting until 70.

Frankly I'm not even sure how I would do this with opensocialsecurity.com as it currently exists, because it assumes only one discount rate and applies it to both spouses in the analysis. A single person could at least iterate.

Cheers.

 

[LadyIJ]

If you are a higher earner, I think the difference (benefit) of waiting is better

 

[big-papa]

If you are a higher earner, I think the difference (benefit) of waiting is better

Analysis of these sorts of scenarios where there is a higher earner and a lower earner almost always end up with the more optimal result being that the higher earner should wait. There are cases, however, where for health reasons or other financial reasons that the right answer is even for the higher earner to start earlier. Most analysis tools can't take these sorts of cases into account, at least not with default settings.

Cheers.

 

[latexman]

The amount I ended up getting was more than I expected to get waiting until 70

Wwwhhhaaattt? A SS mistake, or a lack of understanding? That doesn’t make sense.

 

[Flieger]

Analysis of these sorts of scenarios where there is a higher earner and a lower earner almost always end up with the more optimal result being that the higher earner should wait. There are cases, however, where for health reasons or other financial reasons that the right answer is even for the higher earner to start earlier. Most analysis tools can't take these sorts of cases into account, at least not with default settings.

Cheers.

Yes. By the "numbers" in Open SS or other analysis mechanisms it seems to always say (with disparate earnings), lower earlier, higher later. But as so many point out, there are other factors. History (health), current financial situation, just to name a couple. Each person must make their decision based on their info and what's important (to them). Adding another $50k-$100k or so maximizing the SS part of the equation vs using less of my "stash" now and having more during the Go-Go and Slow-Go years wasn't enough to drive the decision (for me).

Flieger

 

[Markola]

Not always that hard.

Now that they are in their late 80s, both my mother and my father-in-law regret taking SS at 62.

Same for my 87 yo parents, who are surprised to still be alive, thanks to ever-advancing technology.

 

[indiajust]

While the payout distribution can be calculated as a function of lifespan, my understanding of Net Present Value is that it needs to be discounted to a particular date based on assumptions of interest rates and monetary depreciation. Thus, I suggest the tabulated payouts are in nominal dollars, and need a fuzzy, what-if analysis that might be reduced to a delta, inflation minus an interest rate.

The more interesting adjustment that I suggest MUST be part of this NPV calculation is that it needs annuitized, or otherwise adjusted for longevity delta, actual lifespan minus predicted deathdate by FICA. I think the data needs weighted by probability density function, based on this longevity delta.

The final interesting outcome would be to include in the weighting all the negative longevity outcomes, ie to include those who paid in and never got a dime.

It is more math than I care to indulge in, but with a light touch of actuarial effort, and broad guesses at inflation regimes (H/M/L), I would be curious what sort of probabilistic outcomes are revealed, and how that affects decision structure based on personal estimates of HML longevity delta.

 

[junkanoo]

Like the OP said.. the difference is pretty small as a cumulative total if you hit 85. And very few people are thinking about how smart they were because they optimized waiting until 70, at 85. Smaller amount longer, larger amount shorter.

DW just plain WANTED hers at 62. I compromised and went for FRA, when I realized Roth conversions were a waste for us. The amount I ended up getting was more than I expected to get waiting until 70, when I was 55, thanks to those 3 high COLA years in a row. Good enough. It’s pretty hard to find anyone that says they regret taking it when they did, unless they waited until age 70 and got a bad diagnosis a few years later. It is way too easy to think about those ages in your 50s as being normal. When you hit late 60s and 70, the reality of how old 85 is, is way more apparent.

I'm not picking on anyone waiting for FRA to claim. Seems reasonable for many. However, the logic of what you might have thought you would get at age 55, being meaningful, is not upon us. Additionally, everyone benefited from COLA, however that increase benefits those with a higher payout/deferred payout more than, say, your wife. Than again, for most, having marriage be one of compromise beats the alternative.

SS is not about optimization, it's about reasonable decisions given one's situation.

 

[Montecfo]

I think it IS about optimization. But the definition of what that word means can vary from person to person.

My highest level view is that the decision should be based on conditions on the ground.

Presently we are doing the low earner first thing, but no one before FRA as do not want to experience that cut back of spousal benefit.

 

[pb4uski]

While the payout distribution can be calculated as a function of lifespan, my understanding of Net Present Value is that it needs to be discounted to a particular date based on assumptions of interest rates and monetary depreciation. Thus, I suggest the tabulated payouts are in nominal dollars, and need a fuzzy, what-if analysis that might be reduced to a delta, inflation minus an interest rate.

The more interesting adjustment that I suggest MUST be part of this NPV calculation is that it needs annuitized, or otherwise adjusted for longevity delta, actual lifespan minus predicted deathdate by FICA. I think the data needs weighted by probability density function, based on this longevity delta.

The final interesting outcome would be to include in the weighting all the negative longevity outcomes, ie to include those who paid in and never got a dime.

It is more math than I care to indulge in, but with a light touch of actuarial effort, and broad guesses at inflation regimes (H/M/L), I would be curious what sort of probabilistic outcomes are revealed, and how that affects decision structure based on personal estimates of HML longevity delta.

OSS does the above. It calculates expected cash flows, that is, expected SS benefits multiplied by the probability of being alive to receive them based on the mortality tables you select (see screenshot below).

OSS then discunts the expected cash flows to today using the real discount rate that you provide (default is 20 year TIPS rate, currently 2.48%.

Because everything is real $$$ and real discount rates there is no need to make broad guesses at inflation. IOW, all else being equal, nominal cash flows discounted at nominal interest rates are the same as real cash flows disounted at real discount rates.