Draw social security early or not?

[springnr]

Quote:

Originally Posted by NW-Bound I know! And you are the 1st to point out the parallel of "SS delay" and "OMY of work".

Log that to random chance.

[NW-Bound]

Quote:

Originally Posted by ProspectiveBum I'm 42, and DW is 46. Most retirement calculators give us slightly better odds of success if we take SS early...

That may be because you are relatively young.

After I went through the exercise to see how much we would get at 62, FRA, then at 70, out of curiosity I used FIRECalc to see how much we could spend with the various amounts of SS combined with our stash. It turned out that what we could spend (at 100% certainty for 30 years) was highest with SS claimed at FRA. The spendable amount is the same with SS at 62 and 70.

However, the SS at FRA won out over 62 and 70 by less than an amount of 2%, so it is not that significant. What's important was that they all resulted in me able to spend 30% more than what I am spending right now using just the stash.

Quote:

Originally Posted by photoguy ... When I first started seriously planning for ER I had been assuming that SS would be insignificant.

Same here. What a pleasant surprise.

Quote:

Originally Posted by springnr Log that to random chance.

I have always chuckled when seeing posters wanting absolute assurance on anything. Hah, if life could be so predictable and knowable!

But then, if it were, life would be dull.

[Animorph]

Quote:

Originally Posted by Independent I'd say that this Social Security actuaries' table has somewhat higher mortality rates (shorter life expectancies) than most people who ask this question on this site will experience. This SS table is a "total population, period" table. I can think of three adjustments: The table includes everybody, regardless of current health. For example, it includes people who were in hospices on their birthdays, and people who know they have serious diseases. People who ask the "should I defer SS?" question are usually in better health than the perfectly average American (simply because "average" includes those special cases). The table does not distinguish by economic status. Other SS statistics show that people with higher lifetime average incomes tend to outlive those with lower lifetime average incomes. Most posters on this site are in the "higher" income group. A "period" table is a snapshot based on current mortality. It assumes no future improvements in mortality rates. The general trend in mortality rates is downward. The SS actuaries use the "cohort" mortality tables, instead of "period" tables, when they are doing the projections for the SS trust fund.

This is what I used for a personalized life expectancy:

https://www.livingto100.com/

It takes a lot of factors into consideration.

[NW-Bound]

I used a calculator like that (it was mentioned in this forum), and was pleased with the results. And at an annual exam, my doctor declared that my blood tests were "the best he has seen in a long time". BP good, no cholesterol nor excessive glucose, BMI less than 25, no daily pills at all including vitamins, ya da...

A year after that, I was hospitalized for the first time in my life, struggling through major surgeries, wondering how the rest of my life was going to be. No, I was wondering if I was going to make it. And I have scars (many of them) to prove it. I have found out that the phrase YMMV applied to me in a negative way.

[Independent]

Quote:

Originally Posted by photoguy These are good points. Ideally I'd like some sort of regression model (that outputs quantiles) where I could put inputs like race, gender, education, etc. Is this a big effect (say more than 1 year)?

I think you can get a feel for the impact of race and gender from the regular US Life tables: http://www.cdc.gov/nchs/data/dvs/LEWK3_2009.pdf

This is the SS paper on variations by lifetime earnings. See chart 1. Note that it's pretty straight forward to convert annual mortality rates into a probability distribution of ages of death (that's the dx column in a life table).Mortality Differentials by Lifetime Earnings Decile: Implications for Evaluations of Proposed Social Security Law Changes

The cohort vs. period issue is discussed here: Life Tables

Figure 3b. seems to say that a 20 year differential in date of birth converts to about a 3 year difference in life expectancy at age 65. I didn't see a direct comparison (e.g. period at age 62 for calendar year 2012 vs. projected cohort for people born in 1950 on the ssa.gov site, but there may be one).

But, I don't see any discussion of the short term selection effects deriving from people who know they are in bad health. I couldn't find any numbers for annuities that reflect the short term health issue. I did find something on life insurance that's somewhat relevant. http://www.actuary.org/pdf/life/cso2_june01.pdf

The probability that a 62 year-old man will die in the first year after buying a life insurance policy is about 25% of the probability that a 62 year old man who bought his policy 20 years ago will die within the next year (pages A1 and A3, 2.48 vs. 10.28 deaths per 1,000). This is the effect of the insurance company making sure that it doesn't sell policies to people with known health issues. This effect wears off to about 50% five years later.

[Independent]

Quote:

Originally Posted by ProspectiveBum I'm 42, and DW is 46. Most retirement calculators give us slightly better odds of success if we take SS early..

IIRC, a few years ago we had a debate between two posters who seemed to be getting different results from FireCalc. One said that FC told him to defer, the other said that FC said to start at 62.

It turned out the the first person was running FireCalc as a 62 year-old, asking what he should do immediately. The second was younger (maybe 52) asking what he should do 10 years from now. As close as I could guess, it appeared that the order of returns issue gets involved somehow.

So I'd say over the next 16 years the calculators might change their opinion.

[photoguy]

Quote:

Originally Posted by Animorph This is what I used for a personalized life expectancy: https://www.livingto100.com/ It takes a lot of factors into consideration.

Thanks for posting this link as it is quite interesting. However I'm very skeptical of the calculator as I don't think it's based on actual data (i.e. estimated with statistical survival model). It also didn't ask me any questions about race which is one of the biggest covariates. I suspect the calculator is actually asking a bunch of questions about stuff it doesn't use to collect data.

Quote:

Originally Posted by Independent I think you can get a feel for the impact of race and gender from the regular US Life tables: http://www.cdc.gov/nchs/data/dvs/LEWK3_2009.pdf

This is great data. Unfortunately they didn't include asians. sigh.

Quote:

This is the SS paper on variations by lifetime earnings. See chart 1. Note that it's pretty straight forward to convert annual mortality rates into a probability distribution of ages of death (that's the dx column in a life table).Mortality Differentials by Lifetime Earnings Decile: Implications for Evaluations of Proposed Social Security Law Changes

Chart 1 is fascinating (I've attached for forum readers). It shows significant drop in mortality even from deciles 7 to 10 in lifetime earnings. I had known about lower lifespans at the bottom of the income scale but didn't realize it was so large even around the top.

Quote:

Figure 3b. seems to say that a 20 year differential in date of birth converts to about a 3 year difference in life expectancy at age 65. I didn't see a direct comparison (e.g. period at age 62 for calendar year 2012 vs. projected cohort for people born in 1950 on the ssa.gov site, but there may be one).

This is a way bigger effect than I would have thought. Thanks for posting this info.

[NW-Bound]

Quote:

Originally Posted by photoguy ... This is great data. Unfortunately they didn't include asians. sigh. ...

Race is an important hereditary factor, but it can be overwhelmed by others. For example, it is well publicized that Japanese immigrants in the US do not share the same health characteristics of those living in Japan.

About rich people or those of higher socio-economic status having a longer life expectancy, that is no surprise. However, it does not mean one would get longevity by sticking OMY after another in a high-stress job. Surely, a doctor tends to live longer than a labourer, but between an doctor who ER'ed and one that works till 65, the one having more money is not the one healthier.

[haha]

Quote:

Originally Posted by NW-Bound Surely, a doctor tends to live longer than a labourer, but between an doctor who ER'ed and one that works till 65, the one having more money is not the one healthier.

This is pure speculation. May be true, may be false, and I strongly doubt that there is enough quality data to tell which.

[NW-Bound]

OK, I can see now that perhaps the MD that can work till 65 is the one with a stress-free job, such as the one owning the allergy clinic I went to. He most likely outlives the one working in the hospital emergency room, even if the latter retires early.

But I think you know the point that I was trying to make, that there are many secondary factors besides the socio-economic one.

[growing_older]

Quote:

the one that push my decision over the top was what happened to a good friend of mine about 3 years ago and then to my BIL just last year. Both were in very good health "until" they were both diagnosed with pancreatic cancer. Both only lived about 9 months after they were diagnosed. My friend was in his late 50's so he never saw a dime from SS. My BIL was 63 so he collected it for about a year.

I believe that your personal answer to whether it's better for you to start early or late depends on what you are trying to accomplish. There are some cases with spousal differences in ages or benefits that can make a big difference. And there are some cases with minor children that can make a big difference. But if we just think about a simple case of taking SS early or late, the best option depends on your goal.

If your goal is to leave the biggest pile of money behind, then arguments about getting the maximum benefit, using the benefit to offset 401k withdrawals and the like make sense. Similarly concerns about getting any SS at all in case you might die early could also come into play.

In my personal case, none of that matters to me. I have saved enough that I can pay my way at least to age 70 with no SS. Likewise if I die unexpectedly, my heirs will inherit more than enough to set them up comfortably, there's no real need to optimize that since they get more than "enough" regardless. So for me, I do not care about the cases where I die early but collected less SS than if I started earlier.

It seems I need to know my lifespan to know if early or late start to SS is numerically better. But I think that it not relevant for me. If I start SS late, but die earlier than I expected, I will have had enough money to live a lifestyle I was comfortable with, and since I'll be dead, I won't miss the money I could have had if I guessed correctly. But if I turn out to live a long time, I would prefer the larger payout and this is the only case in which it might actually matter to me. So this is the only case I want to optimize for. My plan is to wait as long as possible, although I might file and suspend after FRA just in case I get a poor diagnosis and want to claim as much back dated SS as I can. If you force me to bet on my lifespan, the only side of the bet I want to take is the one where I live a long long time. "Winning" the bet by dying early is an irrelevant case to me (since I'll be dead) so I have no reason to maximize my cash remaining in case I die early. Might be different if there were a spouse or heirs to whom that money might make a difference.

[Jager]

Part of what makes when to draw SS a challenging proposition are the conflicting factors.

Although we often focus on the break-even point and our expected mortality, maximizing our 'take' from Uncle Sam oftentimes misses the larger point.

The retirement years from age 62 to FRA are, for most of us, qualitatively superior to those after FRA. We're younger, we have more energy, our health is better, and we have yet to receive that today-is-the-day-you-die wildcard that some morning we will receive.

Similarly, the retirement years from FRA to age 70 are qualitatively superior to any that will follow.

So the first factor is whether drawing SS is necessary in order to retire in the first place. If it is, then that quickly becomes a powerful argument for taking it sooner, than later. Why trade away those "best" retirement years?

The answer why one might, of course, is written in the second, conflicting, factor: longevity and survivorship risk. If drawing SS early means an essential bet that one or both of a couple will check out fairly early - because the longer term financial horizon looks pitiable - then you hold off drawing SS. You trade away those "best" retirement years hoping not to have to eat dog food in your later years.

For some, SS is rather icing on the cake - they could retire without it. For most, however, it truly is a conundrum.

[W2R]

Quote:

Originally Posted by Jager The retirement years from age 62 to FRA are, for most of us, qualitatively superior to those after FRA. We're younger, we have more energy, our health is better, and we have yet to receive that today-is-the-day-you-die wildcard that some morning we will receive.

As someone who reached FRA this year, I don't wanna hear it!

A week before I turned 66 (my FRA), when weightlifting at the gym one day I lifted a total of 20.5 tons, the most of my entire life. I could never have done that at any younger age. I may never do it again! But I did it.

Yeah, my hair is whiter and I forget things more and my hands are more wrinkled than they once were, but my overall health doesn't seem to be any worse than it was when I retired at age 61.5. It seems better. I am hoping for a few more years, or maybe even another decade before the inevitable steeper decline begins.

While your statement may have been true years ago, when people knew less about how to take care of themselves, I'm not so sure it is true today. It seems to me that in 2014 there are a lot more healthy, active seniors enjoying life and challenging themselves both physically and mentally every day than I recall seeing back in the mid 20th century.

Or maybe that perception reflects the fact that I'm older, now.

[LARS]

Not to turn this into a political discussion, but I am surprised that no one has really mentioned the issue of future changes to SS and how that impacts the decision process. Personally our current thinking is to wait to FRA at a minimum, and then reassess the political SS landscape.

[youbet]

Quote:

Originally Posted by Lancelot most will probably receive a lower total amount of benefits compared with those that delay.

I'm not sure what you mean by "most." 51%? 75%? 99%?

I started at 62 to protect DW due to GPO. But, as a side note, since I began SS just as we began recovery from the "Great Recession," the investment results for the money received so far have been impressive and will likely more than make up for the lower, started early SS checks.

You can't count on investment returns, no argument there. It's just how it's worked out for me.

If you live to a "typical" age and your retirement years have "typical" investment conditions, I don't think it matters very much when you start SS. No doubt unique, individual circumstances may dictate one course of action being better than another for any particular individual/couple.

[youbet]

Quote:

Originally Posted by Jager Part of what makes when to draw SS a challenging proposition are the conflicting factors. Although we often focus on the break-even point and our expected mortality, maximizing our 'take' from Uncle Sam oftentimes misses the larger point. The retirement years from age 62 to FRA are, for most of us, qualitatively superior to those after FRA. We're younger, we have more energy, our health is better, and we have yet to receive that today-is-the-day-you-die wildcard that some morning we will receive. Similarly, the retirement years from FRA to age 70 are qualitatively superior to any that will follow. So the first factor is whether drawing SS is necessary in order to retire in the first place. If it is, then that quickly becomes a powerful argument for taking it sooner, than later. Why trade away those "best" retirement years? The answer why one might, of course, is written in the second, conflicting, factor: longevity and survivorship risk. If drawing SS early means an essential bet that one or both of a couple will check out fairly early - because the longer term financial horizon looks pitiable - then you hold off drawing SS. You trade away those "best" retirement years hoping not to have to eat dog food in your later years. For some, SS is rather icing on the cake - they could retire without it. For most, however, it truly is a conundrum.

I think you're missing something Jeff.......

Your "safe" annual spend rate between 62 and 70 doesn't change much, if at all, whether you start SS at 62, 66 or 70. Try it in FireCalc. You should see little difference in portfolio survival rates whether you plug in your smaller SS at 62 or your larger SS at 70.

See post #62 by NW-Bound (above) for some examples.

There are lots of legitimate and sound reasons to start SS at 62, but having more money to spend while you're younger isn't one of them because it isn't true.

[Jager]

Quote:

Originally Posted by youbet I think you're missing something Jeff....... Your "safe" annual spend rate between 62 and 70 doesn't change much, if at all, whether you start SS at 62, 66 or 70. Try it in FireCalc. You should see little difference in portfolio survival rates whether you plug in your smaller SS at 62 or your larger SS at 70. See post #62 by NW-Bound (above) for some examples. There are lots of legitimate and sound reasons to start SS at 62, but having more money to spend while you're younger isn't one of them because it isn't true.

I wasn't thinking in terms of portfolio survival or safe withdrawal rate. But rather contemplating the question for most people, who don't have such esoteric considerations. For most people, the availability of SS, and when to take it, very much affects when/if they are able to retire.

You're absolutely right that SS-age-of-withdrawal has minimal effect on portfolio survival. But the universe for whom that is an important consideration is, alas, a significant minority.

[Which Roger]

Quote:

Originally Posted by LARS Not to turn this into a political discussion, but I am surprised that no one has really mentioned the issue of future changes to SS and how that impacts the decision process. Personally our current thinking is to wait to FRA at a minimum, and then reassess the political SS landscape.

I don't put too much weight on this in my decision process for the following reasons:

1. If and when SS reform happens, it will likely have a bigger impact on younger people than older people.

2. I'm guessing that however the older folks are impacted, the changes will mainly impact the "wealthy". If I'm "wealthy" when I turn 70, then I will have won the game, and will easily be able to deal with the impact. If I'm struggling at that age, and need every penny of SS, then I won't likely be impacted by the changes.

[haha]

Quote:

Originally Posted by Which Roger I don't put too much weight on this in my decision process for the following reasons: 1. If and when SS reform happens, it will likely have a bigger impact on younger people than older people. 2. I'm guessing that however the older folks are impacted, the changes will mainly impact the "wealthy". If I'm "wealthy" when I turn 70, then I will have won the game, and will easily be able to deal with the impact. If I'm struggling at that age, and need every penny of SS, then I won't likely be impacted by the changes.

What has already been done questions these assumptions. Look at the low income thresholds for Medicare premium increases. This is so far from "wealthy" that it is laughable.
The first step-up comes at $85k for a single taxpayer. That is adjusted AGI, and all the adjustments increase AGI, not decrease it. This amount, and the higher steps are not inflation adjusted.

Ha

[NW-Bound]

I have told this story before, but need it here to lead to what I want to say.

I told another engineer who was 15 years my senior that I wanted to travel when I was young, so that on my deathbed I would not regret not traveling enough. He shook his head and said "Many people on their deathbed just wanted the pain to stop".

About when to take SS, let's take these two extreme suboptimal cases that one may end up in.

1) You took SS at 62, then find out that at 80 you are still going strong. It is highly likely that, due to human nature, you will regret not delaying it so that you would have more now. You will forget all the good stuff or experience you bought with that SS money. You will lament the "wrong choice" to anyone who listens, and tell them to learn from your "mistake".

2) You delay SS till 70, and at 71, find out that you do not have much longer to live. Are you going to regret not taking SS earlier, to spend it on that fancy car, taking that world cruise when you were younger? No, you will be so sad and too busy being scared to think about the unfulfilled bucket list.

In other words, case 1) makes you regret, and in case 2) you don't care. So, with my understanding of human nature as described above, it is probably safer to delay SS to avoid regret.

Caveat: The above assumes that you can live comfortably without SS, and are just debating whether you should take SS to buy more caviar and foie gras. If you are short of funds or if the market tanks, it is far better to take SS to reduce your WR.