Jonathan Clements: About half your retirement accumulation is just the dollars you put in

[nisiprius]

Very interesting, even important article: Show Me the Money. Under reasonable and robust assumptions typical of those usually made in retirement planning, around 40-50% of your final retirement nest egg simply equals the total amount of dollars you put in. This is true whether you calculate in nominal or real dollars.

In one example, he addresses the problem of more contributions coming later in life when they have less time to grow: "What if you don’t get around to saving for retirement for the first 10 years of your career... so you only have 30 years to save for retirement?" However, I strongly suspect this underestimates the effect.

Now, doubling, or a bit more than doubling your savings ain't hay, but it is also not as dramatic as often suggested. To put it in the harshest possible terms, a 2X difference in final retirement nest egg may be the difference between the retirement you want and a somewhat-pinched retirement you have to settle for, but it is not the difference between life and death, particularly if Social Security is in the picture.

Clements says:

That brings us to a perverse conclusion—one I’m almost reluctant to mention: Because savings are so crucial, and because they’re the key driver of your ultimate nest egg, how you invest is somewhat less important.

I think he's right, and I think this has been mentioned in the forum from time to time.

Savings rate is the most important retirement savings decision, not only because of the math but because of the way it drives your financial mindset and habits. The basic raw stock/bond risk decision comes second. And the finer details--index or active, factors or total market, alts or no alts--are a distant third.

Obviously this is not the message that any interested party in the investment industry wants to send. If I may quote a horrible example, this is what people want to believe and this is what the investment industry often insinuates:

A smarter allocation can increase your nest egg by 50 percent. That's a lot easier than increasing your savings by 50 percent.

[AlmstRtrd]

This sounds about right with regard to my retirement numbers when I pulled the plug in 2018. Subtracting significant investment gains from the "final" total shows that about half of what I had at retirement came from savings. Of course, my numbers are very rough and I wasn't a particularly successful investor! One thought... if I have a successful 30-year retirement as I hope, almost all of my gains going forward will be from investments.

[CyclingDuo]

Clements says:

That brings us to a perverse conclusion—one I’m almost reluctant to mention: Because savings are so crucial, and because they’re the key driver of your ultimate nest egg, how you invest is somewhat less important.

I think he's right, and I think this has been mentioned in the forum from time to time.

Savings rate is the most important retirement savings decision, not only because of the math but because of the way it drives your financial mindset and habits. The basic raw stock/bond risk decision comes second. And the finer details--index or active, factors or total market, alts or no alts--are a distant third.

We certainly are on board with that concept - and have posted many times - that your rate of savings is the most important part of the equation.

Clements is not alone...

https://www.marketwatch.com/story/here- ... 2017-07-20

“Clearly, savings seems to trump investing returns for the average American household. This is good news, for saving more is something you actually can control, whereas earning a higher rate of return is infinitesimally more difficult,” Bilello said in a blog post.

https://sparkrental.com/roi-savings-rate-new-study/

A recent analysis by Pension Partners offers some decisive proof that savings rate trumps ROI, demonstrating that discipline matters more than clever investment choices.

[Grt2bOutdoors]

I’ve seen that with my retirement accounts as well. Of course had I speculated in the “right” individual stocks I might have huge multiples of savings or lose it all together. The retirement accounts grow when you add to it, much like a plant will grow when you give it enough water, sun and fertilizer. Deprive it of any one of the three will get you poorer results.

It is not just retirement accounts, same goes for 529 college savings accounts. You generally need to save a lot to see a large accumulation in that account. It’d be great if balances magically grew by leaps and bounds over 18-20 years, but that has not been my experience. It grows when you add money to the account.

[petulant]

This sounds about right with regard to my retirement numbers when I pulled the plug in 2018. Subtracting significant investment gains from the "final" total shows that about half of what I had at retirement came from savings. Of course, my numbers are very rough and I wasn't a particularly successful investor! One thought... if I have a successful 30-year retirement as I hope, almost all of my gains going forward will be from investments.

That last sentence is a key point. While contributions are a significant part of the investor's assets over a lifetime and certainly in the accumulation stage, the relative importance is not static over the investor's lifetime. Myself, I am an early stage accumulator. The vast majority of my balance is still contributions and will continue to be so for several years. I ran an analysis a year or so ago to determine how different investment returns could affect my asset growth over the next 5, 10, etc. years. I found that my returns had a small influence on the asset growth over the next 5-10 years, but after that the importance of returns grew since my contributions would only be a small part of the balance. Generally, for an investor who saved reasonably for 20-30 years with decent returns, contributions in the last few years have a very small effect on asset growth. The investor could possibly stop contributing!

The upshot was that saving more was the key action item for the next 5-10 years, and it mattered little to try to optimize between total stock, S&P 500, international, small-cap, etc. After 2025 or 2030, I will need to take more care.

Occasionally, we have posters begin new threads asking how they should allocate funds. They are often in their 20s or early 30s and have small balances. They are struggling mightily with how to handle international, small-caps, target date funds, etc. I have started letting these investors know the truth: it hardly matters! It matters much more that they actually keep saving! Stick it all in a S&P 500 fund--it really doesn't matter compared to your savings decisions. I think the whole board should adopt this attitude. Tell these starting accumulators to pick a single broad-based stock index fund or target date fund and focus on living life and saving money.

[goodenyou]

The Millionaire Next Door was onto something! LBYM/frugality is the most important ingredient to successful accumulation. I am glad that we have contributors dispelling the idea that investment return and asset allocation are the key to getting to critical mass. The horsepower of accumulation is savings. Return on investing is a promoter.

[SGM]

It is difficult to determine how much of my retirement calculation is the dollars I put into it. Granted we could put in more money later in our working years into both retirement and taxable accounts. 401k limits and IRA limits increased over time and so did income. We also were able to add profit sharing. However, especially in my tax deferred account we bought and sold over the years and current unrealized gains are a small amount compared to realized gains. Unrealized gains are about 40% in a taxable account and 5% in a tax advantaged account (Roth). Realized gains are much higher in the Roth and probably high in the taxable account. I don't see the point in spending much time determining the actual numbers as we have made our most important decisions some time ago.

However, a TIAA-CREF account was established at a first professorship that lasted 4 years. We know exactly how much money was put into the account and the exact value of the account when annuitized at age 66. The dollars put in, including matching, were 3% and the gains were 97%. Initially all the dollars put in including matching by the university were in TIAA traditional. Eventually a little was transferred to a world CREF fund each year per the rules regarding TIAA-CREF exchanges. CREF grew incredibly prior to retirement. Luck had a lot to do with it.

[lack_ey]

Clements makes reasonable assumptions about forward returns, but FWIW historically the returns have been higher, so a greater percentage could have been from investment returns and not contributions. And there could be significant difference based on realized returns over the accumulation phase (getting lucky with 8% real returns over the last decade vs. -3%), or rolling the dice and being heavy in some more speculative investments and/or individual stocks that happen to make a big difference one way or the other.

But yes, in the average scenario that sounds about right, and it could end up being an even higher percentage if returns aren't so cooperative.

If you take another step back, if the target is a dollar amount based on how much spending power you're looking for, then income is another huge driver of what can be saved, in addition to savings rate.

Good and well-taken point about asset allocation details beyond the very basics being way down the list.

[Call_Me_Op]

I can state that over 35 years, my retirement account is about 2X the money I contributed. So at least for me, Jonathan is correct. However, during the 1980's and into mid 1990's, I had very little in stock - which was a mistake.

[Barefoot]

Made me look. I retired end of May last year, and have been withdrawing about 3% since.

My total 30 year contributions are about 22% of my current balance.

I guess I did OK.

[CyclingDuo]

I found that my returns had a small influence on the asset growth over the next 5-10 years, but after that the importance of returns grew since my contributions would only be a small part of the balance. Generally, for an investor who saved reasonably for 20-30 years with decent returns, contributions in the last few years have a very small effect on asset growth. The investor could possibly stop contributing:

I would caution that even if one has been saving consistently for 20-30 years and built up a nice nest egg, when they reach their 50's and continue on into their early 60's there is an opportunity to increase the rate of savings as it usually coincides with the end of funding college educations for children, allows for IRS catch up contributions, the nest is empty, salaries are at or near peak earning years, lifestyle is dialed in and especially for households that are dual income - you can live on one salary and save the other one easier than you could when the kids were in the house (at least that was our experience). Even if your nest egg balance is well established, each additional $240-250K you can add to your nest egg can provide an additional $10K per year in retirement income (4% SWR).

We wouldn't turn down a potential additional $10K, $20K, or $30K annual income in retirement if all it takes is powering on through with our contributions every month during the final decade or so of working years. We would side with continuing contributions and increasing the savings rate provided your lifestyle is comfortable and all of your needs are met. Come January, we will enter our 6th consecutive year of socking it away at a much higher savings rate than any of the prior decades allowed (started saving at least one of our two salaries at our respective ages of 53/56). Now at ages 58/61, we plan to continue doing that during our final years of working.

Moving beyond the actual retirement date, just because households will have pensions, SS, RMD's coming in doesn't mean all of that has to be spent. What is not used or needed after taxes and all household expenses can be saved in spite of the size of the next egg. Obviously, everyone has different goals that may or may not include legacy, charity, self-funding LTC, etc... .

[petulant]

I found that my returns had a small influence on the asset growth over the next 5-10 years, but after that the importance of returns grew since my contributions would only be a small part of the balance. Generally, for an investor who saved reasonably for 20-30 years with decent returns, contributions in the last few years have a very small effect on asset growth. The investor could possibly stop contributing:

I would caution that even if one has been saving consistently for 20-30 years and built up a nice nest egg, when they reach their 50's and continue on into their early 60's there is an opportunity to increase the rate of savings as it usually coincides with the end of funding college educations for children, allows for IRS catch up contributions, the nest is empty, salaries are at or near peak earning years, lifestyle is dialed in and especially for households that are dual income - you can live on one salary and save the other one easier than you could when the kids were in the house (at least that was our experience). Even if your nest egg balance is well established, each additional $240-250K you can add to your nest egg can provide an additional $10K per year in retirement income (4% SWR).

We wouldn't turn down a potential additional $10K, $20K, or $30K annual income in retirement if all it takes is powering on through with our contributions every month during the final decade or so of working years. We would side with continuing contributions and increasing the savings rate provided your lifestyle is comfortable and all of your needs are met. Come January, we will enter our 6th consecutive year of socking it away at a much higher savings rate than any of the prior decades allowed (started saving at least one of our two salaries at our respective ages of 53/56). Now at ages 58/61, we plan to continue doing that during our final years of working.

Moving beyond the actual retirement date, just because households will have pensions, SS, RMD's coming in doesn't mean all of that has to be spent. What is not used or needed after taxes and all household expenses can be saved in spite of the size of the next egg. Obviously, everyone has different goals that may or may not include legacy, charity, self-funding LTC, etc... .

You're right--it might be a good idea to go ahead and keep saving the money. I think a person who has already saved a lot could look into some other options, though, like paying off the house or making some excellent donations, depending on their overall goals and income situation.

[Johnnie]

I've been in that late career "power-on" stage for going on 10 years (but not much longer). One change in the last couple years was to turn-down the "power setting" - new contributions now go into Treasuries as I strive to whittle down equities to a more conservative early-retirement A/A.

I will note that watching the needle rise on that fixed-income balance does reinforce this thesis: It rises quite nicely in power-on contribution mode, even without any investment gains!

[stlrick]

39 years of contributions to a 403(b). The combined employer and employee contributions are 26% of the total accumulation, and that is with a 30:70 allocation since deciding 4-5 years ago that we had won the game.

[smectym]

Hmm, not so sure. We have one small traditional IRA with total contributions of $13,000 and a current value of $58,000. Another small 403(b) from one year as a visiting law school prof 12 years ago, similar numbers. We still get statements on a Janus account we started decades ago with $100—then changed our mind. The number today is several thousand bucks.

Our main accounts, though, may be closer to the model Clements proposes (harder to trace total contributions due to multiple transfers and consolidations). Still, some people at least must be quadrupling, quintupling or better their contributions.

[Maven]

I started following this blog a few weeks ago (thanks to hearing about it via the forum) and I find it to be a perfect way to start my Saturday. I was chatting about this article with my husband this morning. Thanks for posting.

[Portfolio7]

In my early 50's, & the importance of a high savings rate and starting young just overwhelms all the other variables. Every year earlier you start, and every extra percentage is critical. It far outweighs any AA decisions I've made (other than put the money in the market. Keeping it in cash would have been a problem, obv.)

25 years later, I feel at the mercy of the markets, though I think that's less true than it feels. My contributions are a drop in the bucket next to the impact of the market, BUT, still important relative to our financial position 30-40 years from now.

Back to the OP, I've tracked all my contributions so it's really interesting to note how this has changed. A couple data points:

Years investing 1 Contributions 82% + Matching 9% = 91% ; Investment Growth 9%
Years investing 10 Contributions 61% + Matching 13% = 74% ; Investment Growth 26%
Years investing 23 Contributions 27% + Matching 18% = 45% ; Investment Growth 55%

From which it becomes clear how critical those early decisions are, and also points to a weakness of looking at percentages. If we look into the future, I suspect that after 40 years of accumulative investing your original contributions are likely going to be well under 25% of your total. However, the magnitude of the total depends on your early savings rate. Also, a note that my matching is a little odd due to a unique benefit.

[mrc]

Spouse's TIAA is an exception to that rule. She contributed in the 1980s through about 1991.

Employee contribution: $3377
Employer contribution: $6442

Value in Feb 2017 was $296,415.

Clearly, the time horizon matters!

[Portfolio7]

Clearly, the time horizon matters!

+1. I tried to make a similar point in my post as well.

[firebirdparts]

obviously a circular argument can be made. Based on my brilliant assumptions/study/dataset, some people didn’t grow their money much. Therefore, growth is unimportant.

The SP500 in total return has quadrupled in ten years. Do we not care? Sometimes these articles and posts on this forum have made up phrases like “most” or “very few” followed by some made up data about lagging reasonable investing for some reason or another.

[Vanguard Fan 1367]

My wife bought 20k worth of Fidelity Magellan in 1994. She paid the 3 percent load. She has added nothing to that account. It is in an IRA and is now worth about 150k and we have changed the holding to bonds because of Bogleheads advice to keep bonds in tax deferred accounts.

Our overall retirement accumulation is more like what Clements says.

Someone has said that 1994 was a good time to get into the market.

[willthrill81]

Someone that contributes the $6k max to an IRA over 30 years at a 5% return will have $398,633, of which $180k, 45% of the final balance, is their contributions.

But I don't really see why it matters. It certainly isn't discouraging to me. If anything, it reinforces to me the importance of investing early and often. Those that saved but didn't invest in the above scenario would have 45% as much as had they gotten a 5% return.

In our own circumstance, I expect that upon retirement, slightly over 50% of our portfolio will be contributions.

[Silence Dogood]

Savings rate is the most important retirement savings decision, not only because of the math but because of the way it drives your financial mindset and habits. The basic raw stock/bond risk decision comes second. And the finer details--index or active, factors or total market, alts or no alts--are a distant third.

Perhaps this is why I don't fret too much about all of these other issues.

Some people want a fine-tuned portfolio that is personalized to their situation. I'm satisfied with a Vanguard Target Retirement fund.

With my savings rate, I'm pretty sure I'll do fine either way.

[White Coat Investor]

I've been saving and investing for about 15 years on a rising income and am financially independent by any reasonable measurement. Of my retirement portfolio, 72% represents savings and 28% represents earnings.

The longer it takes you to get to where you're going, the more is likely to be portfolio gains rather than savings. If you get there in 10 years, it's probably mostly savings. If you get there in 40, it's probably mostly gains.

[willthrill81]

The longer it takes you to get to where you're going, the more is likely to be portfolio gains rather than savings. If you get there in 10 years, it's probably mostly savings. If you get there in 40, it's probably mostly gains.

That's something that the FIRE crowd has going in their favor: the brute force of a high savings rate can fairly easily overcome mediocre returns.

[goodenyou]

I've been saving and investing for about 15 years on a rising income and am financially independent by any reasonable measurement. Of my retirement portfolio, 72% represents savings and 28% represents earnings.

The longer it takes you to get to where you're going, the more is likely to be portfolio gains rather than savings. If you get there in 10 years, it's probably mostly savings. If you get there in 40, it's probably mostly gains.

High savings/earning ratios favor high earning individuals. Plus, when you are high income earner, your need to take risk is reduced. Conservative investing doctors fit this mold. I am 70/30 saving/investing returns myself.

[MathIsMyWayr]

A simple math: Save the often recommended 15% of earnings for 30 years.
2 (return of 100%) * 30 (years) * 15% (savings rate) = 9x: Does it look close to the Fidelity recommended number?
If you up the savings rate from 15% to 20%, then you will end up with 12x the final earnings.

[telemark]

Since we like to make a point here of being careful about tenses, the title should probably read

About half your retirement accumulation may be just the dollars you put in

if you are starting now and Mr. Clements' prediction of 4% real growth turns out to be accurate. People who started earlier have often done much better than that.

[usagi]

High savings/earning ratios favor high earning individuals.

The system favors those who are willing to create a high savings to earning ration. That is achieved by living below your means and there is tremendous amount of variance in standards of living and the cost associated with achieving a given standard.

[goodenyou]

High savings/earning ratios favor high earning individuals.

The system favors those who are willing to create a high savings to earning ration. That is achieved by living below your means and there is tremendous amount of variance in standards of living and the cost associated with achieving a given standard.

You got it. You can live very very well spending a mid-6 figure sum per year, and retire very very well while saving a mid-6 figure sum per year at the same time. It doesn’t work out too well when you spend most of a high 6-figure income. The more money I made as I aged, the less I’ve wanted. High savings rates coupled with the willingness to live at a fraction of your earned income in retirement allows for early retirement and a high savings/earnings ratio.

[Canopus]

For TIAA-CREF participants, if one logs onto the website and clicks on Account Summary > Retirement> Contributions, the resulting page will display:

) Total Balance
) Contributions from you
) Contributions from your employer.

Easy enough to determine what percentage of your total balance has come from contributions as compared to investment returns.

Canopus

[JBTX]

At 5% real return, and investment will double in around 15 years. A person who contributions evenly over 30 years has an average "duration" of contribution of 15 years, so the 50% savings/ 50% earnings scenario seems plausible. If you were invested mostly in stocks the last 30 years you may have done better than that.

[wootwoot]

High savings/earning ratios favor high earning individuals.

The system favors those who are willing to create a high savings to earning ration. That is achieved by living below your means and there is tremendous amount of variance in standards of living and the cost associated with achieving a given standard.

From the data presented someone with high earnings and a low savings ratio will also be favored.

[JBTX]

I was curious and looked at my quicken. I have 20 years of full data and 10 years before that was only partial. Looking at current net worth, and looking at savings over 30 years, including estimates of 1st 10 years of savings, it looks like at the end of 30 years about 1/3 savings and 2/3 earnings.

[Phineas J. Whoopee]

Sure. I don't know why anyone would be surprised by that.

Some people will [hyperbolic sports metaphor]. Others will [hyperbolic metaphor about clothing].

Barring those, early in one's investing career nearly all increase in one's portfolio will be from new saving. Eventually total return may overtake it, but one is still saving if one still can and desires to.

PJW

[CULater]

Spouse's TIAA is an exception to that rule. She contributed in the 1980s through about 1991.

Employee contribution: $3377
Employer contribution: $6442

Value in Feb 2017 was $296,415.

Clearly, the time horizon matters!

Maybe not as much as you think when you adjust for inflation.

I think the time horizon that matters the most is the one from when you retire to when you kick.

[leonidas]

I just looked online and its 47% contributions and 53% returns for me since 1994. I have about 10-12 more years till retirement.

[Portfolio7]

Spouse's TIAA is an exception to that rule. She contributed in the 1980s through about 1991.

Employee contribution: $3377
Employer contribution: $6442

Value in Feb 2017 was $296,415.

Clearly, the time horizon matters!

Maybe not as much as you think when you adjust for inflation.

I think the time horizon that matters the most is the one from when you retire to when you kick.

Investment returns are still roughly 90% after adjusting for constant currency. I agree with the original sentiment, Time Horizon matters.

As for time after retirement, that's a pretty individual choice, I think.

[CULater]

Spouse's TIAA is an exception to that rule. She contributed in the 1980s through about 1991.

Employee contribution: $3377
Employer contribution: $6442

Value in Feb 2017 was $296,415.

Clearly, the time horizon matters!

Maybe not as much as you think when you adjust for inflation.

I think the time horizon that matters the most is the one from when you retire to when you kick.

Investment returns are still roughly 90% after adjusting for constant currency. I agree with the original sentiment, Time Horizon matters.

As for time after retirement, that's a pretty individual choice, I think.

I don't want to be a Doubting Thomas, but if you had put $9,819 into an investment in 1981, in order for it to grow to almost $300,000 by 2017 it would have to generate about a 9.4% return compounded annually for 38 years. But the entire amount wasn't even invested in 1981. I'm pretty sure TIAA- fixed didn't get that. Just how did an investment of less than $10K made in the 1980s do that?

[LilyFleur]

Returns have been above 90%.
Most bogleheads do not take that kind of risk.
Now in a conservative BH-type portfolio and happily living a fairly modest retirement lifestyle.

[Artsdoctor]

If I remember correctly, DFA did a nice study years ago looking at several things influencing total return over an investor's lifetime. The amount put in was a very significant factor, if not the most significant factor for the majority of investors. However, timing was also incredibly important. Depending on market forces at the beginning, the middle, and the end of your investment career, your final balances of a half-dozen investors putting in the exact same amount can vary wildly at the end. This can easily be illustrated if you look at investors having an initial (and sole) investment of $1M at varying entry dates over several decades; each 40-year period of investment will have varying amounts. So I wouldn't be surprised if we get a lot of different results on this thread, simply depending on the investment amounts at varying investment dates--especially if people have been investing for decades.

[22twain]

When I retired three years ago, after contributing to my 403(b) plan for 32 years, always 50% TIAA Traditional and 50% CREF Stock, contributions were about 43% of the total accumulation. With continued growth and no withdrawals since then, the figure is now about 35%.

[vitaflo]

This is something I noticed a while back and was a bit disheartening to be honest. For me, 20 years in, it's 65% from savings and 35% from returns. My purchases 20 years ago have of course more than doubled, but it's also a very small amount relative to what I invest every month now. $150/mo purchases in 2001 being worth $400 now doesn't seem like a big deal when I can put away $2k per month now.

In essence, the larger purchases I've made as I've gotten older just haven't had as much time to compound. Not much more to it than that.

[vitaflo]

Sure. I don't know why anyone would be surprised by that.

Because most articles on compound interest explore the incredible multiplicative gains you get over time, leading people to believe that they are going to end up with 3x, 4x, 5x of the money they put in. But this is usually only true of the initial deposit, not the aggregate total of savings. I can see why people may be surprised by it (even if after learning the truth it seems fairly obvious).

[Hydromod]

This is something I noticed a while back and was a bit disheartening to be honest. For me, 20 years in, it's 65% from savings and 35% from returns. My purchases 20 years ago have of course more than doubled, but it's also a very small amount relative to what I invest every month now. $150/mo purchases in 2001 being worth $400 now doesn't seem like a big deal when I can put away $2k per month now.

In essence, the larger purchases I've made as I've gotten older just haven't had as much time to compound. Not much more to it than that.

Interesting. I calculate that, 14 years in, my contributions are 58%. That's with me dumping in the max the whole time, including makeup contributions over the last 8 years. Asset allocation at work?

[Day9]

This makes me think I should browse the "Personal Finance (Not Investing)" and "Personal Consumer Issues" boards more than this "Investing - Theory, News & General" board.

[nedsaid]

It seems that I have contributed one dollar for every three in my portfolio. One factor is that I started investing back in 1984, just about when one of the most powerful bull market in history began.

[vitaflo]

This is something I noticed a while back and was a bit disheartening to be honest. For me, 20 years in, it's 65% from savings and 35% from returns. My purchases 20 years ago have of course more than doubled, but it's also a very small amount relative to what I invest every month now. $150/mo purchases in 2001 being worth $400 now doesn't seem like a big deal when I can put away $2k per month now.

In essence, the larger purchases I've made as I've gotten older just haven't had as much time to compound. Not much more to it than that.

Interesting. I calculate that, 14 years in, my contributions are 58%. That's with me dumping in the max the whole time, including makeup contributions over the last 8 years. Asset allocation at work?

Most likely sequence of returns, or in this case contributions. I started investing in 2000, not a great time to start. And most of my contributions came in the last 8 years, which, while it's been a great time to be in the market, is only 8 years.

[Portfolio7]

Spouse's TIAA is an exception to that rule. She contributed in the 1980s through about 1991.

Employee contribution: $3377
Employer contribution: $6442

Value in Feb 2017 was $296,415.

Clearly, the time horizon matters!

Maybe not as much as you think when you adjust for inflation.

I think the time horizon that matters the most is the one from when you retire to when you kick.

Investment returns are still roughly 90% after adjusting for constant currency. I agree with the original sentiment, Time Horizon matters.

As for time after retirement, that's a pretty individual choice, I think.

I don't want to be a Doubting Thomas, but if you had put $9,819 into an investment in 1981, in order for it to grow to almost $300,000 by 2017 it would have to generate about a 9.4% return compounded annually for 38 years. But the entire amount wasn't even invested in 1981. I'm pretty sure TIAA- fixed didn't get that. Just how did an investment of less than $10K made in the 1980s do that?

If you invested $9,819 60/40 in 1981 you'd have $360K today. To get a better view, an 80/20 portfolio of $9,819 invested lump sum in 1985 (to split the contribution years) yieldsed $290K. I think if you assume $800-900 per year for 11 or 12 years from 1980 to 1991 the numbers are at least as good, I estimated that a 65/35 portfolio would have made it to $296K. There were some very good stretches of returns in that time period. The number seem credible to me.

Even if they were made up, the projections I've done suggest that anyone investing for 40 years could see 90% of total savings in investment returns. It might also be much lower, depending on how aggressively or defensively the individual invested.

[grayfox]

This sounded interesting, so I calculated my numbers. I started saving and investing c. 1980. Unfortunately, I only have data going back to Dec-1992, so I can't calculate how much was contribution vs. gain before then. I will assume that the entire balance in Dec-1992 was contribution, i.e. zero return from c. 1980 to Dec-1992.

When I stopped contributing, Net-Contribution/Balance was 34.84%
Added: If I assume 50% of Dec-1992 balance was contribution, Net-Contribution/Balance was 25.30%. This is probably better a estimate of the actual percent.

By Sep-2015, Net-Contribution/Balance was 0%, i.e. I had taken out everything I put in.
Sep-2019, Net-Contribution/Balance = -5%. i.e. have taken out more than I put in.

So far, I have put in $41.70 and taken out $52.61. What a racket!
Again, this assumes the entire balance in Dec-1992 was contribution, but probably a lot (half?) was actually gain.

Added: I am guessing that the range might be more like 25% to 75% depending on circumstances, rather than 50% for everyone.