Realistic rate of return the next 20 years
-
- Posts: 55
- Joined: Sun Jan 23, 2011 8:10 pm
Realistic rate of return the next 20 years
I realize this is a HUGE crystal ball question, but assuming inflation is in the 3-4% range over the next 20 years, do you feel it is safe to assume the overall stock market will produce 6% returns to investors over that time period?
:roll:
:roll:
-
- Posts: 55
- Joined: Sun Jan 23, 2011 8:10 pm
-
- Posts: 132
- Joined: Sat Jan 01, 2011 2:35 pm
- Opponent Process
- Posts: 5157
- Joined: Tue Sep 18, 2007 9:19 pm
-
- Posts: 838
- Joined: Wed Dec 22, 2010 12:28 pm
I always use 7% rate of return and 3% annual inflation when projecting the growth of my investments during my accumulation phase, which I predict will extend for the next 33 years.Thetightfist wrote:Maybe I should put it this way..........when you consider how much is needed for retirement, what rate of return do you plug into your calculations, erring a bit on the conservative side?
-
- Posts: 461
- Joined: Sat Sep 11, 2010 9:01 am
I think it depends on what you mean by "safe to assume." It seems reasonable and somewhat conservative to use 2-3% real as a 20-year expected return. But this just means that you assume that there is an even chance you'll do better and an even chance you'll do worse. How much worse might it be?do you feel it is safe to assume the overall stock market will produce 6% returns to investors over that time period?
According to a calculator on Moneychimp that I cannot link to, the standard deviation (SD) for 20-year real returns (for U.S. stocks) has been 3.1%. That means a 20-year real return of negative 1% to 0% would only be one SD from an expected return of 2-3%. Whether or not this is a normal distribution, one SD seems reasonably likely. Even 2 SDs doesn't seem out of the question: that would be a return of negative 3-4% real.
Suppose we were to define as "pretty safe" an assumption that 20-year actual returns will be no more than 2 SDs from the expected return. That would suggest, starting with a 2-3% expected real return, it is pretty safe to assume that the actual real return will be between -4% and 9%. If we start with a slightly higher expected return--say, 4% real--then it would be pretty safe to assume an actual real return of between -2% and 10%.
I guess my point is this: with stocks, the return you expect and the return it is safe to rely on are two different things. I personally would not feel safe relying on any positive return at all for stocks over the next 20 years, though of course I do expect a significant positive return.
-
- Posts: 12073
- Joined: Fri Sep 18, 2009 1:10 am
- Adrian Nenu
- Posts: 5228
- Joined: Thu Apr 12, 2007 6:27 pm
When no one knows for sure, you diversify with stocks and bonds. If stocks have huge returns you still make some money but if they have low returns or losses you are not wiped out. That's how diversification works.I realize this is a HUGE crystal ball question, but assuming inflation is in the 3-4% range over the next 20 years, do you feel it is safe to assume the overall stock market will produce 6% returns to investors over that time period?
Adrian
anenu@tampabay.rr.com
Depends on your asset mix. I'm about 70/30 and I typically assume around 4% real with a std. dev. of about 10%.
That means 95% of the time I expect yearly returns to be -16% to 24% after factoring out inflation.
If you aren't using std. deviation and calculating risk of failure, your retirement calculations could be seriously flawed. I recommend this calculator that will run monte carlo simulations.
http://www.flexibleretirementplanner.co ... RPWeb.html
That means 95% of the time I expect yearly returns to be -16% to 24% after factoring out inflation.
If you aren't using std. deviation and calculating risk of failure, your retirement calculations could be seriously flawed. I recommend this calculator that will run monte carlo simulations.
http://www.flexibleretirementplanner.co ... RPWeb.html
Dan,dandan14 wrote:Depends on your asset mix. I'm about 70/30 and I typically assume around 4% real with a std. dev. of about 10%.
That means 95% of the time I expect yearly returns to be -16% to 24% after factoring out inflation.
If you aren't using std. deviation and calculating risk of failure, your retirement calculations could be seriously flawed. I recommend this calculator that will run monte carlo simulations.
http://www.flexibleretirementplanner.co ... RPWeb.html
I think one of the most interesting parts of this simulator, is that the likelihood of reaching my target goals change very little if I take a below average risk or aggressive approach. It really gives me a sense that the need for an aggressive allocation is less when saving is adequate... which has been stated many times by the senior posters here...
Anuj
For very optimistic guesses, see J Seigel, who will give a guess based on earnings yield (1 / PE) or based on historical returns. If earnings are booming, use PE7 or PE10 instead, or somehow "normalize" them based on recent history. I don't think earnings are extremely high today.
For pessimistic guesses, see the GMO paper* linked somewhere at this forum, which uses a good methodology but fudges downwards because they are just pessimistic people.
Or W Bernstien, who adds about 1% or if you're lucky, 2% to dividends, using the Gordon equation, seemingly ignoring net share buybacks.
* this one was not authored by Grantham... maybe Inker?
For pessimistic guesses, see the GMO paper* linked somewhere at this forum, which uses a good methodology but fudges downwards because they are just pessimistic people.
Or W Bernstien, who adds about 1% or if you're lucky, 2% to dividends, using the Gordon equation, seemingly ignoring net share buybacks.
* this one was not authored by Grantham... maybe Inker?
- Rick Ferri
- Posts: 9708
- Joined: Mon Feb 26, 2007 10:40 am
- Location: Georgetown, TX. Twitter: @Rick_Ferri
- Contact:
Here's my forecast for the next 30 years:
The Portfolio Solutions 30-Year Market Forecast
Rick Ferri
The Portfolio Solutions 30-Year Market Forecast
Rick Ferri
The Education of an Index Investor: born in darkness, finds indexing enlightenment, overcomplicates everything, embraces simplicity.
- White Coat Investor
- Posts: 17413
- Joined: Fri Mar 02, 2007 8:11 pm
- Location: Greatest Snow On Earth
Rick's numbers make sense to me. People who think stocks will only return 2% real over 10-30 years are overly pessimistic in my opinion. My best guess is my slice and diced 75/25 portfolio will have a 5% real return over my investing horizon. Currently, over 7 years through one severe bear, it is 7.5% nominal. Inflation in that time period has been 2.5%/year, leaving me with 5% real. Seems like my personal historical experience matches my personal theory. Those who are overly optimistic about stock returns will overallocate to stocks while those who are overly pessimistic will underallocate, and pay the price in lower returns.Rick Ferri wrote:Here's my forecast for the next 30 years:
The Portfolio Solutions 30-Year Market Forecast
Rick Ferri
Ask yourself this-if you really thought stocks were only going to return 2 or 3% real, why would you put your money into them? TIPS will guarantee you those kinds of returns. Barring severe unexpected inflation, nominal treasuries and long-term CDs are likely to give you similar returns. Why run market risk?
1) Invest you must 2) Time is your friend 3) Impulse is your enemy |
4) Basic arithmetic works 5) Stick to simplicity 6) Stay the course
- White Coat Investor
- Posts: 17413
- Joined: Fri Mar 02, 2007 8:11 pm
- Location: Greatest Snow On Earth
An assumption of 0% real return is likely to result in a bunch of oversaving, working too long, and not enjoying your time and money as much as you could.jimkinny wrote:The only safe thing to do is assume 0% real return and any thing above that will allow you to achieve your goals sooner. IMO if you are counting on 4%, you are gambling. Control what you can control: income, spending and savings (investments).
Take an average joe. Let's say he makes $60,000 a year and saves 10% of it. At 0% real, after 40 years, when it is time to retire, he will have $6K*40=$240K. He may need that money for say 35 years, so at 0% real, the maximum possible withdrawal rate is 1/35th, assuming no fluctuations, no inheritance left behind etc. That's $6857/year. That'll replace a grand total of 11% of his pre-retirement income. That's not going to go very far. Assuming he wants his nest egg to replace 60% of his pre-retirement income (we'll assume SS makes up whatever else he needs) he needs to be putting away $31,500, a savings rate of 53%. That's an awful lot of deferred spending that one would incur by using your 0% real assumption. How sad to scrimp and save and deny yourself your entire life and then die with millions in the bank.
1) Invest you must 2) Time is your friend 3) Impulse is your enemy |
4) Basic arithmetic works 5) Stick to simplicity 6) Stay the course
Eye opening isn't it. If you can still meet goals by taking less risk, then its an obvious choice!p_qrs_t wrote:Dan,dandan14 wrote:Depends on your asset mix. I'm about 70/30 and I typically assume around 4% real with a std. dev. of about 10%.
That means 95% of the time I expect yearly returns to be -16% to 24% after factoring out inflation.
If you aren't using std. deviation and calculating risk of failure, your retirement calculations could be seriously flawed. I recommend this calculator that will run monte carlo simulations.
http://www.flexibleretirementplanner.co ... RPWeb.html
I think one of the most interesting parts of this simulator, is that the likelihood of reaching my target goals change very little if I take a below average risk or aggressive approach. It really gives me a sense that the need for an aggressive allocation is less when saving is adequate... which has been stated many times by the senior posters here...
Anuj
-
- Posts: 6972
- Joined: Fri Feb 23, 2007 6:46 pm
- Location: Allentown–Bethlehem–Easton, PA-NJ Metropolitan Statistical Area
Thanks Rick. It's few that put their "opinions" in a site that is used for public viewing, in a commercial manner (I hope you're correct. Heck, in less than 30 years, I'll be "dust").Rick Ferri wrote:Here's my forecast for the next 30 years:
The Portfolio Solutions 30-Year Market Forecast
Rick Ferri
- Ron
PS: USAF still rules :lol: (could you expect less, from me? Remember, we don't need the Navy to transport us to where the action is...)
Last edited by Ron on Wed Jan 26, 2011 12:03 pm, edited 1 time in total.
-
- Posts: 12073
- Joined: Fri Sep 18, 2009 1:10 am
I've used Rick's numbers in the past as well as GMO numbers (they tend to be more pessimistic though) and use 5% real for my expected return. That's perhaps on the aggressive side but plugging a few numbers from Rick's page into my allocation, I still am at 5% (actually 5.105%).
Part of my process is to track my returns against a baseline (i.e., how much would I have if I had gotten 5% real every year). This started in 1998 - I am somewhat behind but perhaps less than you might think given the way the market has gone. The realization is that if I continue to lag my baseline, my options are: save more, work longer or lower income expectations in retirement.
It is very much a moving target but the exercise itself is where the value lies in my opinion.
Ken
Part of my process is to track my returns against a baseline (i.e., how much would I have if I had gotten 5% real every year). This started in 1998 - I am somewhat behind but perhaps less than you might think given the way the market has gone. The realization is that if I continue to lag my baseline, my options are: save more, work longer or lower income expectations in retirement.
It is very much a moving target but the exercise itself is where the value lies in my opinion.
Ken
EmergDoc wrote:An assumption of 0% real return is likely to result in a bunch of oversaving, working too long, and not enjoying your time and money as much as you could.jimkinny wrote:The only safe thing to do is assume 0% real return and any thing above that will allow you to achieve your goals sooner. IMO if you are counting on 4%, you are gambling. Control what you can control: income, spending and savings (investments).
Take an average joe. Let's say he makes $60,000 a year and saves 10% of it. At 0% real, after 40 years, when it is time to retire, he will have $6K*40=$240K. He may need that money for say 35 years, so at 0% real, the maximum possible withdrawal rate is 1/35th, assuming no fluctuations, no inheritance left behind etc. That's $6857/year. That'll replace a grand total of 11% of his pre-retirement income. That's not going to go very far. Assuming he wants his nest egg to replace 60% of his pre-retirement income (we'll assume SS makes up whatever else he needs) he needs to be putting away $31,500, a savings rate of 53%. That's an awful lot of deferred spending that one would incur by using your 0% real assumption. How sad to scrimp and save and deny yourself your entire life and then die with millions in the bank.
I really do not think we disagree that much. My central point was to assume the worst, plan for the worst and if things work out for the better, then one is ahead of one's plan. To die with millions in the bank is indeed foolish. Life is precious and should be enjoyed. If you find that you are accumulating more than anticipated, then that certainly would be a reason to change your plan. But to assume something about the future that we really can not predict, seems to me to be unnecessary.
This is a bit like assuming a withdrawal rate in retirement of 4% is going to always be sustainable, which quite likely is nonsense. It may work out that way but I would want to be in a position that I am not counting on that rate. So, put yourself in the best position that you can so that if your rate of return is not what you would expect, based upon historical data, then you will still be okay and not dependent upon others to take care of you.
-
- Posts: 1264
- Joined: Wed Oct 07, 2009 9:11 am
I plug in 5% real, or 4% real if I am erring on the conservative side. That's my best guess, emphasis on guess.Thetightfist wrote:Maybe I should put it this way..........when you consider how much is needed for retirement, what rate of return do you plug into your calculations, erring a bit on the conservative side?
I'm not sure what is safe to assume for a 20 year time period, probably nothing. I just save whatever I can without impacting my quality of life.
- Rick Ferri
- Posts: 9708
- Joined: Mon Feb 26, 2007 10:40 am
- Location: Georgetown, TX. Twitter: @Rick_Ferri
- Contact:
Yes, USAF rules those 12,000 foot runways because they need lot of room to land, unlike Navy/Marine carrier pilots who can land on a dime.Ron wrote:Rick,USAF still rules :lol: (could you expect less, from me? Remember, we don't need the Navy to transport us to where the action is...)
Rick Ferri
The Education of an Index Investor: born in darkness, finds indexing enlightenment, overcomplicates everything, embraces simplicity.
-
- Posts: 461
- Joined: Sat Sep 11, 2010 9:01 am
That's 4.1% real after inflation AND average taxes AND fees. Somebody investing in a low-cost index fund would have fared much better. Of course, index funds didn't exist for most of that period so perhaps 4.1% is a fair number.beareconomy wrote:if you look at mymoneyblog.com, there was an article a few days ago about the S and P returning real since about 1920 of 4.1% real after inflation. I thought that was interesting in order to compare to TIPS.
but it is a long walk to the nearest pub after a long day on the job.Rick Ferri wrote:Yes, USAF rules those 12,000 foot runways because they need lot of room to land, unlike Navy/Marine carrier pilots who can land on a dime.Ron wrote:Rick,USAF still rules :lol: (could you expect less, from me? Remember, we don't need the Navy to transport us to where the action is...)
Rick Ferri
Don't trust me, look it up. https://www.irs.gov/forms-instructions-and-publications
Returns come from dividends, earnings (or dividend) growth and change in p/e.beareconomy wrote:if you look at mymoneyblog.com, there was an article a few days ago about the S and P returning real since about 1920 of 4.1% real after inflation. I thought that was interesting in order to compare to TIPS.
The dividend rate on Vanguard's total stock market fund is 1.5% at the moment. Dividends were much higher 90 years ago.
To get above 4% real growth, you need earnings of the companies you own (not the economy as a whole) to grow at more than 2.5% real per year. That would be above historical rates. Alternatively, p/e could keep expanding indefinitely.
For those who expect higher than 4% equity returns, do you expect higher than 2.5% real earnings growth or ever growing p/e ratios? If neither, how do you expect to get higher returns?
30 year TIPS are now yielding about 2% real. If you expect higher returns, please explain what you know that isn't reflected in the market.
The world is a safer place than it was in 1920. That should mean lower returns. Investing costs are lower, so investors don't need as high gross returns to end up with the same net return. That should also mean lower returns.
- White Coat Investor
- Posts: 17413
- Joined: Fri Mar 02, 2007 8:11 pm
- Location: Greatest Snow On Earth
4% is a lot different than 0%. I think 4% is a reasonable guess. I can retire on 4%. But planning on 0% is just silly.richard wrote:Returns come from dividends, earnings (or dividend) growth and change in p/e.beareconomy wrote:if you look at mymoneyblog.com, there was an article a few days ago about the S and P returning real since about 1920 of 4.1% real after inflation. I thought that was interesting in order to compare to TIPS.
The dividend rate on Vanguard's total stock market fund is 1.5% at the moment. Dividends were much higher 90 years ago.
To get above 4% real growth, you need earnings of the companies you own (not the economy as a whole) to grow at more than 2.5% real per year. That would be above historical rates. Alternatively, p/e could keep expanding indefinitely.
For those who expect higher than 4% equity returns, do you expect higher than 2.5% real earnings growth or ever growing p/e ratios? If neither, how do you expect to get higher returns?
30 year TIPS are now yielding about 2% real. If you expect higher returns, please explain what you know that isn't reflected in the market.
The world is a safer place than it was in 1920. That should mean lower returns. Investing costs are lower, so investors don't need as high gross returns to end up with the same net return. That should also mean lower returns.
If TSM returns 4%, and you believe EM/Small Value/microcap etc has a premium over it, say a 2% premium then your overall equity return could be 5%+ without requiring any change in PE, earnings growth, or increase in dividends.
1) Invest you must 2) Time is your friend 3) Impulse is your enemy |
4) Basic arithmetic works 5) Stick to simplicity 6) Stay the course
Long term stocks might have returned 7% real, so maybe that is a fair middle of the road assumption. On the other hand many developed markets had smaller returns and current P/E10, Tobin's q are on the high side which implies a somewhat lower return going forward. So maybe 6% middle of the road is better, maybe 4% is on the low side but not terribly low. The money you have invested today may in fact do worse, but hopefully you are adding more month by month so some will end up invested in a more attractive environment.
Something like 2% real for bonds historically, but they are very high priced, so better to think in terms of 1% real.
So, 60/40 would be about .6*6% + .4*1% = 4% real, or if we get the lower range in stocks, more like 3%.
Sure, maybe you get a little more from some sort of MPT benefit, so add 0.1%
If you want to assume 3% inflation we are talking about 6%-7% nominal.
I use 3% inflation and 5% nominal for my "lousy" case and 6% for my "middle of the road" case, and 7% for my "pretty good case", recognizing that I actually have no clue and it could be a lot worse.
When I want lots of meaningless digits I use one of my home grown Monte Carlo simulations. (though truth be told while writing and studying the behavior of MCS was a lot of fun when I started, I have not run one in a long time as it is actually a waste of time as the errors are so huge the benefits beyond pencil and paper don't really exist).
Given my age I am more concerned about 20 years out, though might hope to live another 40 years.
It is entirely possible that returns will be horrible; if you need a lot more than bond rates you need to increase savings rate, IMHO. You cannot increase likelihood of a good outcome by changing assumptions in your spread sheet, and you cannot count on working into your 70s or even late 60s.
Something like 2% real for bonds historically, but they are very high priced, so better to think in terms of 1% real.
So, 60/40 would be about .6*6% + .4*1% = 4% real, or if we get the lower range in stocks, more like 3%.
Sure, maybe you get a little more from some sort of MPT benefit, so add 0.1%
If you want to assume 3% inflation we are talking about 6%-7% nominal.
I use 3% inflation and 5% nominal for my "lousy" case and 6% for my "middle of the road" case, and 7% for my "pretty good case", recognizing that I actually have no clue and it could be a lot worse.
When I want lots of meaningless digits I use one of my home grown Monte Carlo simulations. (though truth be told while writing and studying the behavior of MCS was a lot of fun when I started, I have not run one in a long time as it is actually a waste of time as the errors are so huge the benefits beyond pencil and paper don't really exist).
Given my age I am more concerned about 20 years out, though might hope to live another 40 years.
It is entirely possible that returns will be horrible; if you need a lot more than bond rates you need to increase savings rate, IMHO. You cannot increase likelihood of a good outcome by changing assumptions in your spread sheet, and you cannot count on working into your 70s or even late 60s.
We live a world with knowledge of the future markets has less than one significant figure. And people will still and always demand answers to three significant digits.
0% seems overly conservative.EmergDoc wrote:4% is a lot different than 0%. I think 4% is a reasonable guess. I can retire on 4%. But planning on 0% is just silly.
I just took a look at Shiller's data. From 1871 to 2010, real earnings increased at 1.6%/year. Add the 1.5% current dividend rate and you get 3% real.
You may get a higher return by taking more risk. On the other hand, if risk means anything, it means the chance of lower returns.
That seems to assume growth in price is independent of dividend. I see no reason why company A who sends profits out in dividends and nearly identical company B who buys back stocks, or who reinvests internally should have the same price increase: in the later case the increase in price will be higher, commensurate with the money for stock buy backs and internal investments. Return is the sum of price increase and dividends, and should be the same in either case.richard wrote:Returns come from dividends, earnings (or dividend) growth and change in p/e.beareconomy wrote:if you look at mymoneyblog.com, there was an article a few days ago about the S and P returning real since about 1920 of 4.1% real after inflation. I thought that was interesting in order to compare to TIPS.
The dividend rate on Vanguard's total stock market fund is 1.5% at the moment. Dividends were much higher 90 years ago.
To get above 4% real growth, you need earnings of the companies you own (not the economy as a whole) to grow at more than 2.5% real per year. That would be above historical rates. Alternatively, p/e could keep expanding indefinitely.
For those who expect higher than 4% equity returns, do you expect higher than 2.5% real earnings growth or ever growing p/e ratios? If neither, how do you expect to get higher returns?
30 year TIPS are now yielding about 2% real. If you expect higher returns, please explain what you know that isn't reflected in the market.
The world is a safer place than it was in 1920. That should mean lower returns. Investing costs are lower, so investors don't need as high gross returns to end up with the same net return. That should also mean lower returns.
In the past companies were more likely to return profits via one mechanism, now by another mechanism, who cares? If they are fundamentally generating the same profits company A and B will have the same total return.
Seems to me.
We live a world with knowledge of the future markets has less than one significant figure. And people will still and always demand answers to three significant digits.
The data do not seem to be on your side. This paper by Arnott and Asness found that lower dividend rates have resulted in lower growth rates rather than the higher rate you suggest: http://papers.ssrn.com/sol3/papers.cfm? ... _id=295974Rodc wrote:That seems to assume growth in price is independent of dividend. I see no reason why company A who sends profits out in dividends and nearly identical company B who buys back stocks, or who reinvests internally should have the same price increase: in the later case the increase in price will be higher, commensurate with the money for stock buy backs and internal investments. Return is the sum of price increase and dividends, and should be the same in either case.
In the past companies were more likely to return profits via one mechanism, now by another mechanism, who cares? If they are fundamentally generating the same profits company A and B will have the same total return.
Seems to me.
Forgive me if I'm belaboring the obvious but I think there are two questions here that are very important to distinguish clearly:
1. What return do you expect for stocks over the next 20 years?
2. What return are you relying on for stocks over the next 20 years?
They are both important questions, and many of the answers here have been to 1., which is fine. The OP may have been asking 2. though. And surely, given the variability of 20-year stock returns, the answers to these 2 questions should not be the same.
I agree that it would be strange to expect a 0% 20-year real return for stocks. Why would I invest in stocks if I expected no real return? I don't, however, see why it is foolish to rely on nothing more than a 0% real return. That does mean saving at a rate that, in retrospect, will probably turn out to be unnecessarily high. The higher savings rate is essentially insurance against a bad, improbable outcome. Similarly, the life insurance that I recently purchased will also probably, in retrospect, prove to have been unnecessary.
Consider another analogy inspired by a recent thread by Nisiprius. Let the variability of the 20-year real return of stocks be represented by the roll of 2 6-sided dice, with a roll of "snake eyes" representing 0% real return. When you roll the dice, you don't expect to roll snake eyes--just a 1 in 36 chance. But do you want to rely on--i.e., bet your retirement or freedom from poverty on--not rolling snake eyes? If not, then do you really think a 20-year real return for stocks of 0% is that much less probable than rolling snake eyes?
1. What return do you expect for stocks over the next 20 years?
2. What return are you relying on for stocks over the next 20 years?
They are both important questions, and many of the answers here have been to 1., which is fine. The OP may have been asking 2. though. And surely, given the variability of 20-year stock returns, the answers to these 2 questions should not be the same.
I agree that it would be strange to expect a 0% 20-year real return for stocks. Why would I invest in stocks if I expected no real return? I don't, however, see why it is foolish to rely on nothing more than a 0% real return. That does mean saving at a rate that, in retrospect, will probably turn out to be unnecessarily high. The higher savings rate is essentially insurance against a bad, improbable outcome. Similarly, the life insurance that I recently purchased will also probably, in retrospect, prove to have been unnecessary.
Consider another analogy inspired by a recent thread by Nisiprius. Let the variability of the 20-year real return of stocks be represented by the roll of 2 6-sided dice, with a roll of "snake eyes" representing 0% real return. When you roll the dice, you don't expect to roll snake eyes--just a 1 in 36 chance. But do you want to rely on--i.e., bet your retirement or freedom from poverty on--not rolling snake eyes? If not, then do you really think a 20-year real return for stocks of 0% is that much less probable than rolling snake eyes?
Last edited by DeGrief on Sun Jan 30, 2011 1:08 am, edited 1 time in total.
Rate for planning purposes
I assume 3% real for a ~70/30 portfolio over a 20 year horizon.
There are a lot of valid and interesting points being raised for discussion, but at the end of the day you need to plant a flag for planning purposes to avoid paralysis by analysis.
I love this forum, but some of the threads remind me of a trip to the paint store with my wife. She can never pick a color because she can find flaws in all of them.
There are a lot of valid and interesting points being raised for discussion, but at the end of the day you need to plant a flag for planning purposes to avoid paralysis by analysis.
I love this forum, but some of the threads remind me of a trip to the paint store with my wife. She can never pick a color because she can find flaws in all of them.
I think jack bogle has stated he expects returns to be in the 6% range.
I really think the only way it can be less is if the American public refuses to curb entitlements and we simply start printing money.
Other than that, assuming a return over this time span of 0-3% seems pessimistic. In essence you are betting against companies being profitable, which is hard to justify over the long term.
I really think the only way it can be less is if the American public refuses to curb entitlements and we simply start printing money.
Other than that, assuming a return over this time span of 0-3% seems pessimistic. In essence you are betting against companies being profitable, which is hard to justify over the long term.
"get out and live, you are dead an awfully long time" - Jimmy Demaret
I think there are two questions that have to be considered.richard wrote:Returns come from dividends, earnings (or dividend) growth and change in p/e.
The dividend rate on Vanguard's total stock market fund is 1.5% at the moment. Dividends were much higher 90 years ago.
To get above 4% real growth, you need earnings of the companies you own (not the economy as a whole) to grow at more than 2.5% real per year. That would be above historical rates. Alternatively, p/e could keep expanding indefinitely.
For those who expect higher than 4% equity returns, do you expect higher than 2.5% real earnings growth or ever growing p/e ratios? If neither, how do you expect to get higher returns?
First, are we talking about real returns from this exact moment in time forward or are we talking about real returns over an entire period of 30, 40 or 50 years? From this exact moment in time, your 4% real estimate for TSM makes sense to me. But the market is currently at a relatively higher valuation than it was, say, two years ago. From that point in time, when dividends were closer to 3%, you might calculate real returns as 5.5% real going forward. So - I would say my 5% real is based on an entire period, not this particular moment in time.
Second, are there asset classes that have higher risk and therefore higher expected returns than Total Stock Market? I would say that yes, there are. Emerging Markets, Small Value, Value in general, REIT (perhaps), even Vanguard Developed Markets Index is currently yielding closer to 3%. So, I would say that is another way to achieve better than 4% real.
I'm assuming 3.5% average annualized inflation and 7% equity returns for long time periods (>= 30 years). I consider that to be sufficiently conservative. The lowest 30 year annualized return for the S&P 500 TR from 1928 through 2009 was 8.0% (1929-1958), and 8.5% for 40 years (1929-1968). I admit the S&P 500 is not a perfect proxy for the Total Stock Market, but it at least gives some historical perspective.
But what about the next 20 years (original poster's question)? I'm less comfortable with 7% as being sufficently conservative! Using the S&P 500 TR from 1928 through 2009 as a historical guide, the lowest 20 year return was 2.4% annualized from 1929-1948. The 10th percentile for 20 year returns was 6.8% annualized. A 7% return is certainly more likely than not. But given that "Past Performance is Not Indicative of Future Returns" (SEC), and that Black Swans like to poop on your retirement plans, I'd want a better margin of safety for 20 year (or shorter) time periods.
But what about the next 20 years (original poster's question)? I'm less comfortable with 7% as being sufficently conservative! Using the S&P 500 TR from 1928 through 2009 as a historical guide, the lowest 20 year return was 2.4% annualized from 1929-1948. The 10th percentile for 20 year returns was 6.8% annualized. A 7% return is certainly more likely than not. But given that "Past Performance is Not Indicative of Future Returns" (SEC), and that Black Swans like to poop on your retirement plans, I'd want a better margin of safety for 20 year (or shorter) time periods.
To what extent do recent dividends reflect the great recession? If they are low due to companies hording cash or something, does the dividend method of estimation hold (for example Gordon Equation)?
All methods I know of seem to have some major massive assumptions. I mentioned some for the use of dividends. One paper says maybe I should not worry about those concerns. Time will tell.
It seems unclear to me that any of the estimation methods are worth much, or even better than just looking at global historical ranges.
As to what to plan for (rather than what one might expect), to be safe one probably ought to figure that 2% real (ie sort of bond rates) really is a possibility. No one wants to hear that of course; but that does not make an invalid position. Risk is real: there is no guarantee that stocks over the next 20 or even 30 years will beat treasuries.
All methods I know of seem to have some major massive assumptions. I mentioned some for the use of dividends. One paper says maybe I should not worry about those concerns. Time will tell.
It seems unclear to me that any of the estimation methods are worth much, or even better than just looking at global historical ranges.
As to what to plan for (rather than what one might expect), to be safe one probably ought to figure that 2% real (ie sort of bond rates) really is a possibility. No one wants to hear that of course; but that does not make an invalid position. Risk is real: there is no guarantee that stocks over the next 20 or even 30 years will beat treasuries.
We live a world with knowledge of the future markets has less than one significant figure. And people will still and always demand answers to three significant digits.
I agree with DeGrief's point. This thread got me to try some retirement projection calculators. I could scrape by with 0% real return (I'm saving like crazy at the moment), so that's worth knowing. On the other hand I invest aggressively, so I know the expected return is somewhat higher, and it might be higher still or it might be negative. I don't know what's going to happen so I just choose a reasonable AA.
Another point is, you can just watch how things go and adjust accordingly. Even though I could scrape by with 0% real return I should invest aggressively (for a while) and I'll see where that puts me 5 or 10 years from now (with a horizon somewhat longer still to come).
edit: typo
Another point is, you can just watch how things go and adjust accordingly. Even though I could scrape by with 0% real return I should invest aggressively (for a while) and I'll see where that puts me 5 or 10 years from now (with a horizon somewhat longer still to come).
edit: typo
Last edited by 555 on Thu Jan 27, 2011 8:45 pm, edited 1 time in total.
- fluffyistaken
- Posts: 1435
- Joined: Fri Apr 04, 2008 1:32 pm
1927-2010 Large Cap Stocks- ~6.5% real
1927-2010 5 Year Treasuries- ~2% real
1927-2010 50/50- ~5% real
I'm hoping for 7.5% real return from my 60/40 Swedroesque portfolio but we are over saving so that even negative real returns would be enough to retire by 62.
1927-2010 5 Year Treasuries- ~2% real
1927-2010 50/50- ~5% real
I'm hoping for 7.5% real return from my 60/40 Swedroesque portfolio but we are over saving so that even negative real returns would be enough to retire by 62.
Last edited by snodog on Thu Jan 27, 2011 11:59 am, edited 4 times in total.
One other issue with your earlier comment. You appear to have made a subtle analytical shift.Rodc wrote:All methods I know of seem to have some major massive assumptions. I mentioned some for the use of dividends. One paper says maybe I should not worry about those concerns. Time will tell.
The comparison I was making (and the scenario the paper mainly analyzes) is the dividend and growth rates for all companies at time 1 compared to the aggregates at time 2. You r response compared company A, with a low dividend rate, to company B, with a high rate. These are two different issues and intuitions that apply to one case don't necessarily apply to the other case.
I think this is probably the right way to look at it - an expected return with its corresponding standard deviation. That way you will have a nice range of potential asset value at the end of the period to plan appropriately (for example, plan for 1/2 std deviation below expected value or whatever you think is appropriate).dandan14 wrote:Depends on your asset mix. I'm about 70/30 and I typically assume around 4% real with a std. dev. of about 10%.
That means 95% of the time I expect yearly returns to be -16% to 24% after factoring out inflation.
If you aren't using std. deviation and calculating risk of failure, your retirement calculations could be seriously flawed. I recommend this calculator that will run monte carlo simulations.
http://www.flexibleretirementplanner.co ... RPWeb.html
However, I would be really interested in seeing what people would use for the expected value (discussed above) and, more importantly, the standard deviation and why?
Dandan - why did you choose 10% for std dev?
Good post...DeGrief wrote:Forgive me if I'm belaboring the obvious but I think there are two questions here that are very important to distinguish clearly:
1. What return do you expect for stocks over the next 20 years?
2. What return are you relying on for stocks over the next 20 years?
They are both important questions, and many of the answers here have been to 1., which is fine. The OP may have been asking 2. though. And surely, given the variability of 20-year stock returns, the answers to these 2 questions should not be the same.
I agree that it would be strange to expect a 0% 20-year real return for stocks. Why would I invest in stocks if I expected no real return? I don't, however, see why it is foolish to rely on nothing more than a 0% real return. That does mean saving at a rate that, in retrospect, will probably turn out to be unnecessarily high. The higher savings rate is essentially insurance against a bad, improbable outcome. Similarly, the life insurance that I recently purchased will also probably, in retrospect, prove to have been unnecessary.
Consider another analogy inspired by a recent thread by Nisiprius. Let the variability of the 20-year real return of stocks be represented by the roll of 2 6-sided die, with a roll of "snake eyes" representing 0% real return. When you roll the dice, you don't expect to roll snake eyes--just a 1 in 36 chance. But do you want to rely on--i.e., bet your retirement or freedom from poverty on--not rolling snake eyes? If not, then do you really think a 20-year real return for stocks of 0% is that much less probable than rolling snake eyes?
I actually have 3 data points in mind...
I expect to see 5-6% real....
I plan on about 3-4% real
It's possible I'll only see 1%-2% real
Planning for lower than expected is a very good idea... But you can't use the lowest POSSIBLE outcome as a plan... If we only see 1%-2%, I work until I drop or (more likely), I retire with a minimal lifestyle.
I figure I can retire (with my wanted lifestyle) at 60-62 on my planned rate.... I think it's very likely I'll get to retire earlier than that....
But I'm saving hard assuming the lower rate... Saving too much is never a problem... just means I get to retire early...