Argument for a 100% stock allocation for 20 yr horizon

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
User avatar
Topic Author
fundtalker123
Posts: 901
Joined: Tue Feb 27, 2007 3:18 am

Argument for a 100% stock allocation for 20 yr horizon

Post by fundtalker123 »

Question: In past 20 year periods how "risky" was it to be 100% stock vs. 50/50 stock/bond?

I took data on stock and bond returns from 1928-2011 posted here: http://pages.stern.nyu.edu/~adamodar/Ne ... stret.html
I calculated real returns by subtracting annual rates of inflation posted here: http://inflationdata.com/inflation/infl ... ation.aspx

I calculated that in about 8% of 20 year periods 100% stock had lower total real returns than 50/50

Does that mean 100% stock was too risky? Well, if 50/50 yielded you the minimum return one needed, then one would have been taking an 8% chance of having unacceptably lower return.

However, how could one have known at the start of a 20 year period that 50/50 was going to yield the minimum return needed? There is no specific return guaranteed by 50/50.

By different reasoning one could look at 100% being less risky than 50/50.

I calculated that in 0% of 20 year periods 100% stock had a more than 13% lower total return than 50/50, while in 90% of 20 year periods it had >13% higher return than 50/50, in 50% of periods it had >50% higher return, and in 30% of periods it had >100% higher return.

If doing no worse than 13% lower total return than offered by 50/50 (a return which is unknown to begin with) were acceptable, than 100% stocks would have offered this while offering a very high chance of a substantially greater return. Or, in other words, 50/50 had an unknown chance to provide an acceptable total return, while 100% stock had a 92% chance to offer a higher return and only an 8% chance to offer a lower return, with the worst case being a 13% lower return.

Doesn't 100% stock sound pretty good compared to 50/50 by this analysis?

(assuming one has the stomach to hold 100% stocks, and assuming I didn't miscalculate any numbers)
TheEternalVortex
Posts: 2576
Joined: Tue Feb 27, 2007 8:17 pm
Location: San Jose, CA

Re: Argument for a 100% stock allocation for 20 yr horizon

Post by TheEternalVortex »

There are less then 5 non-overlapping 20 year periods in your sample. How confident can you be based on that?
chaz
Posts: 13604
Joined: Tue Feb 27, 2007 1:44 pm

Re: Argument for a 100% stock allocation for 20 yr horizon

Post by chaz »

Risk = reward.
Chaz | | “Money is better than poverty, if only for financial reasons." Woody Allen | | http://www.bogleheads.org/wiki/index.php/Main_Page
User avatar
Topic Author
fundtalker123
Posts: 901
Joined: Tue Feb 27, 2007 3:18 am

Re: Argument for a 100% stock allocation for 20 yr horizon

Post by fundtalker123 »

TheEternalVortex wrote:There are less then 5 non-overlapping 20 year periods in your sample. How confident can you be based on that?
How confident can I be about what? When you invest, if you do not base your asset allocation on past history of returns of bonds vs. stocks, what do you base it on?
User avatar
nisiprius
Advisory Board
Posts: 52216
Joined: Thu Jul 26, 2007 9:33 am
Location: The terrestrial, globular, planetary hunk of matter, flattened at the poles, is my abode.--O. Henry

Re: Argument for a 100% stock allocation for 20 yr horizon

Post by nisiprius »

In athletics, we take it for granted that new records are being set all the time, and that just because something hasn't happened before doesn't mean it will never happen.

In statistics, we learn that you can get good estimates of the average with relatively reasonable amounts of data, but that it takes humongous amounts of data to get good estimates of the amount of variability, and mega-humongous amounts to estimate the higher-order moments. My experience is limited but I don't think I've ever seen anyone calculate an estimated kurtosis (the fatness of the tails), and I don't know where to find a table to get the 5% significance limits on the value you calculate.

Everywhere, we learn that there are "fat tails" and that our ability to predict low-probability, high consequences events is extremely poor. Over and over again, investors crash and burn and complain that the chances of what actually happened was so small that it shouldn't have happened. The collapse of LTCi was claimed to have been a "ten-sigma event."

It is somewhat understandable that people treat 1929 as if it were something that "shouldn't count," because it has almost past beyond living memory, but people have already forgotten how surprising and "impossible" 2008-2009 seemed to be when it was happening.

In every edition of Stocks for the Long Run through the most recent in 2007, Jeremy Siegel has included some variant of the statement
never in any of the past 175 years would a buyer of newly-issued 30-year bonds have outperformed an investor in a diversified portfolio of common stocks held over the same period.
True for 175 years until it wasn't. I am really hoping he will write a new edition pretty soon as I'm very interested to see how he's going to phrase this next time around. What, never? No, never. What, never? Well, hardly ever.
Annual income twenty pounds, annual expenditure nineteen nineteen and six, result happiness; Annual income twenty pounds, annual expenditure twenty pounds ought and six, result misery.
hazlitt777
Posts: 1033
Joined: Fri Aug 12, 2011 4:10 am
Location: Wisconsin

Re: Argument for a 100% stock allocation for 20 yr horizon

Post by hazlitt777 »

fundtalker123 wrote:Question: In past 20 year periods how "risky" was it to be 100% stock vs. 50/50 stock/bond?

I took data on stock and bond returns from 1928-2011 posted here: http://pages.stern.nyu.edu/~adamodar/Ne ... stret.html
I calculated real returns by subtracting annual rates of inflation posted here: http://inflationdata.com/inflation/infl ... ation.aspx

I calculated that in about 8% of 20 year periods 100% stock had lower total real returns than 50/50

Does that mean 100% stock was too risky? Well, if 50/50 yielded you the minimum return one needed, then one would have been taking an 8% chance of having unacceptably lower return.

However, how could one have known at the start of a 20 year period that 50/50 was going to yield the minimum return needed? There is no specific return guaranteed by 50/50.

By different reasoning one could look at 100% being less risky than 50/50.

I calculated that in 0% of 20 year periods 100% stock had a more than 13% lower total return than 50/50, while in 90% of 20 year periods it had >13% higher return than 50/50, in 50% of periods it had >50% higher return, and in 30% of periods it had >100% higher return.

If doing no worse than 13% lower total return than offered by 50/50 (a return which is unknown to begin with) were acceptable, than 100% stocks would have offered this while offering a very high chance of a substantially greater return. Or, in other words, 50/50 had an unknown chance to provide an acceptable total return, while 100% stock had a 92% chance to offer a higher return and only an 8% chance to offer a lower return, with the worst case being a 13% lower return.

Doesn't 100% stock sound pretty good compared to 50/50 by this analysis?

(assuming one has the stomach to hold 100% stocks, and assuming I didn't miscalculate any numbers)
Roger Gibson, in his book Asset Allocation, made a point I never forgot. He talked about one of the dangers of a well diversified portfolio: If it underperforms the S&P 500 there can be the temptation for an investor to bail out and chase preformance. He wrote the book with financial advisor's in mind. He wanted them to be aware of the fact that their client's psychology had to be taken into account when advising people. They in effect, had to be educated, so that if a well diversified portfolio underperformed they wouldn't get all upset and make a bad decision which could well result in selling low and buying high.

In theory, I suppose one could argue that all stocks portfolio might be reasonable, in the abstract. But could you, me or anybody really stomach the volatility? I think the temptation to bail from a total stock portfolio would be much higher at times than from a 50/50 stock bond allocation, or a 33,33,33 stock, bond ,gold portfolio like I am in. Why? Because we can tolerate underperformance that is positive much better than underperformance that actually entails losses.

It is one thing to talk academically or in the abstract, but to actually have to live with the volatility...that is what you would have to think real seriously about.
User avatar
Topic Author
fundtalker123
Posts: 901
Joined: Tue Feb 27, 2007 3:18 am

Re: Argument for a 100% stock allocation for 20 yr horizon

Post by fundtalker123 »

nisiprius wrote:In athletics, we take it for granted that new records are being set all the time, and that just because something hasn't happened before doesn't mean it will never happen.
So you would argue that because 50/50 could theoretically have an arbitrarily higher total return (maybe even 1000% higher) than 100% stocks over a 20 year period, even though 50/50 has historically only returned up to at most 13% higher return less than 8% of the time, it is better to be 50/50 than 100/0?
User avatar
nisiprius
Advisory Board
Posts: 52216
Joined: Thu Jul 26, 2007 9:33 am
Location: The terrestrial, globular, planetary hunk of matter, flattened at the poles, is my abode.--O. Henry

Re: Argument for a 100% stock allocation for 20 yr horizon

Post by nisiprius »

fundtalker123 wrote:
nisiprius wrote:In athletics, we take it for granted that new records are being set all the time, and that just because something hasn't happened before doesn't mean it will never happen.
So you would argue that because 50/50 could theoretically have an arbitrarily higher total return (maybe even 1000% higher) than 100% stocks over a 20 year period, even though 50/50 has historically only returned up to at most 13% higher return less than 8% of the time, it is better to be 50/50 than 100/0?
No, because I'm not interested in getting the highest possible average total return--if I were, I wouldn't stop at 100% stocks, I'd use leverage.

On the other hand, I'm very interested in not exceeding my risk tolerance, because exceeding one's risk tolerance is a good way to do really serious financial harm to ones' self. On the evidence of postings in this forum, it seems to me that a) people typically make more mistakes in the direction of underestimating the risk of stocks than overestimating it, and b) people typically make more mistakes overestimating their personal risk tolerance than underestimating it.

What I am saying is that the past is the only thing we have to go on, but that the past is a particularly unreliable guide to predicting extreme events. So the fact that the range of things that happened over 20-year holding periods... during the period 1926 to the present... in the United States... didn't include anything unacceptable to you, isn't really enough to go by. My point is not that bonds once beat stocks. My point is that something that, as of 2007, "has never happened in 175 years," actually happened just a couple of years later.
Annual income twenty pounds, annual expenditure nineteen nineteen and six, result happiness; Annual income twenty pounds, annual expenditure twenty pounds ought and six, result misery.
User avatar
Topic Author
fundtalker123
Posts: 901
Joined: Tue Feb 27, 2007 3:18 am

Re: Argument for a 100% stock allocation for 20 yr horizon

Post by fundtalker123 »

nisiprius wrote:
fundtalker123 wrote:
nisiprius wrote:In athletics, we take it for granted that new records are being set all the time, and that just because something hasn't happened before doesn't mean it will never happen.
So you would argue that because 50/50 could theoretically have an arbitrarily higher total return (maybe even 1000% higher) than 100% stocks over a 20 year period, even though 50/50 has historically only returned up to at most 13% higher return less than 8% of the time, it is better to be 50/50 than 100/0?
No, because I'm not interested in getting the highest possible average total return--if I were, I wouldn't stop at 100% stocks, I'd use leverage.

On the other hand, I'm very interested in not exceeding my risk tolerance, because exceeding one's risk tolerance is a good way to do really serious financial harm to ones' self. On the evidence of postings in this forum, it seems to me that a) people typically make more mistakes in the direction of underestimating the risk of stocks than overestimating it, and b) people typically make more mistakes overestimating their personal risk tolerance than underestimating it.

What I am saying is that the past is the only thing we have to go on, but that the past is a particularly unreliable guide to predicting extreme events. So the fact that the range of things that happened over 20-year holding periods... during the period 1926 to the present... in the United States... didn't include anything unacceptable to you, isn't really enough to go by. My point is not that bonds once beat stocks. My point is that something that, as of 2007, "has never happened in 175 years," actually happened just a couple of years later.
At the end of the day you still have to decide asset allocations, so how do you decide?
User avatar
Topic Author
fundtalker123
Posts: 901
Joined: Tue Feb 27, 2007 3:18 am

Re: Argument for a 100% stock allocation for 20 yr horizon

Post by fundtalker123 »

hazlitt777 wrote:
In theory, I suppose one could argue that all stocks portfolio might be reasonable, in the abstract. But could you, me or anybody really stomach the volatility? I think the temptation to bail from a total stock portfolio would be much higher at times than from a 50/50 stock bond allocation, or a 33,33,33 stock, bond ,gold portfolio like I am in. Why? Because we can tolerate underperformance that is positive much better than underperformance that actually entails losses.

It is one thing to talk academically or in the abstract, but to actually have to live with the volatility...that is what you would have to think real seriously about.
In past stock market downturns I found I had no trouble holding on to the stock I had, but I had trouble getting myself to rebalance bonds to stocks. If I was 100% stock I wouldn't have had that problem. (although maybe one reason I had no trouble holding on to the stock I had was that I knew I had a certain amount of bond that I wouldn't lose).
Mortgasm
Posts: 285
Joined: Mon Feb 06, 2012 6:54 pm

Re: Argument for a 100% stock allocation for 20 yr horizon

Post by Mortgasm »

You only have a 20 year horizon for one year. Then it's 19, then 18 then 17...

Are you still going to be 100% in stocks at 2 years from your target date?

It's a straw man argument. No one has a static allocation. Most people get more conservative as you get closer to the target date. 100% is fine at year 20, sure. But it's not at year 5 because you want to preserve your wins.
User avatar
Topic Author
fundtalker123
Posts: 901
Joined: Tue Feb 27, 2007 3:18 am

Re: Argument for a 100% stock allocation for 20 yr horizon

Post by fundtalker123 »

Mortgasm wrote:You only have a 20 year horizon for one year. Then it's 19, then 18 then 17...

Are you still going to be 100% in stocks at 2 years from your target date?

It's a straw man argument. No one has a static allocation. Most people get more conservative as you get closer to the target date. 100% is fine at year 20, sure. But it's not at year 5 because you want to preserve your wins.
My calculation concerns the past total returns of hold 100/0 vs. 50/50 for 20 year periods. The result is you most often came out far ahead, except 8% of the time you got between a 0 and 13% lower return.
Mortgasm
Posts: 285
Joined: Mon Feb 06, 2012 6:54 pm

Re: Argument for a 100% stock allocation for 20 yr horizon

Post by Mortgasm »

I think you haven't considered what a 20 year horizon means. it means you really need the money (or some of it) beginning in year 21. If you don't need the money, and don't care if you suffer big losses in year 19, then your horizon is longer.
mwm158
Posts: 475
Joined: Sat Jun 25, 2011 6:55 pm

Re: Argument for a 100% stock allocation for 20 yr horizon

Post by mwm158 »

deleted
Last edited by mwm158 on Wed Jan 07, 2015 8:04 pm, edited 1 time in total.
User avatar
Topic Author
fundtalker123
Posts: 901
Joined: Tue Feb 27, 2007 3:18 am

Re: Argument for a 100% stock allocation for 20 yr horizon

Post by fundtalker123 »

Mortgasm wrote:I think you haven't considered what a 20 year horizon means. it means you really need the money (or some of it) beginning in year 21. If you don't need the money, and don't care if you suffer big losses in year 19, then your horizon is longer.
Historically it didn't matter if you suffered big losses in year 19, because the total return after 20 years was almost always still higher for 100/0 than 50/50 even including instances where there was a big loss in year 19. In the worst case, starting one particular year out of 64 overlapping 20 year periods, it was only 13% lower. In >90% of cases it was higher.
Last edited by fundtalker123 on Fri Mar 16, 2012 11:16 pm, edited 1 time in total.
User avatar
Topic Author
fundtalker123
Posts: 901
Joined: Tue Feb 27, 2007 3:18 am

Re: Argument for a 100% stock allocation for 20 yr horizon

Post by fundtalker123 »

mwm158 wrote:In my several years and several million hands of poker, I went to a showdown and lost to a royal flush exactly twice. Both times I had a boat. The evil part about it is they happened on 2 consecutive nights. The odds of that happening were as close to zero as you can get. I had a million hands worth of historical data telling me this wouldn't happen once, let alone twice. How many sigmas is that? Anyway, my point is that 100 years worth of stock data means absolutely diddly nothing. I wouldn't be 100% in anything.
So your theory for asset allocation is, I don't know anything about what might happen for asset class A or asset class B, so I'll just choose 50/50?
mwm158
Posts: 475
Joined: Sat Jun 25, 2011 6:55 pm

Re: Argument for a 100% stock allocation for 20 yr horizon

Post by mwm158 »

deleted
Last edited by mwm158 on Wed Jan 07, 2015 8:04 pm, edited 1 time in total.
TheEternalVortex
Posts: 2576
Joined: Tue Feb 27, 2007 8:17 pm
Location: San Jose, CA

Re: Argument for a 100% stock allocation for 20 yr horizon

Post by TheEternalVortex »

fundtalker123 wrote:
TheEternalVortex wrote:There are less then 5 non-overlapping 20 year periods in your sample. How confident can you be based on that?
How confident can I be about what? When you invest, if you do not base your asset allocation on past history of returns of bonds vs. stocks, what do you base it on?
Do you realize the error in concluding that "8%" of the time the 50/50 wins when you only actually have 4.5 such periods? How confident can you be that 100% is a good bet based on 4.5 samples? You don't have 60 different samples because they are all highly correlated with each other.

For example, I randomly generated a list of 60 coin flips (0 or 1). It so happens that in this set, over 92% of sets of 20 consecutive coinflips have more 1s than 0s. Can we conclude anything from this?

1,0,1,1,0,0,1,0,1,0,0,1,1,0,1,1,0,0,1,1,1,0,0,1,1,1,1,1,1,1,0,1,0,0,1,0,0,0,0,1,1,1,1,0,0,1,1,1,0,1,1,0,1,1,0,0,0,0,0,0

(Overall there are only slightly more 1s than 0s: 32 vs 28)
letsgobobby
Posts: 12073
Joined: Fri Sep 18, 2009 1:10 am

Re: Argument for a 100% stock allocation for 20 yr horizon

Post by letsgobobby »

fundtalker123 wrote:
mwm158 wrote:In my several years and several million hands of poker, I went to a showdown and lost to a royal flush exactly twice. Both times I had a boat. The evil part about it is they happened on 2 consecutive nights. The odds of that happening were as close to zero as you can get. I had a million hands worth of historical data telling me this wouldn't happen once, let alone twice. How many sigmas is that? Anyway, my point is that 100 years worth of stock data means absolutely diddly nothing. I wouldn't be 100% in anything.
So your theory for asset allocation is, I don't know anything about what might happen for asset class A or asset class B, so I'll just choose 50/50?
Not necessarily, but what about 80/20? Many would consider that reasonable if you have a 20+ year horizon.
User avatar
Kevin M
Posts: 15789
Joined: Mon Jun 29, 2009 3:24 pm
Contact:

Re: Argument for a 100% stock allocation for 20 yr horizon

Post by Kevin M »

fundtalker123: What is your AA? Has this analysis caused you to reconsider your AA?

Kevin
If I make a calculation error, #Cruncher probably will let me know.
User avatar
nisiprius
Advisory Board
Posts: 52216
Joined: Thu Jul 26, 2007 9:33 am
Location: The terrestrial, globular, planetary hunk of matter, flattened at the poles, is my abode.--O. Henry

Re: Argument for a 100% stock allocation for 20 yr horizon

Post by nisiprius »

fundtalker, your analysis isn't new or unknown to Bogleheads.

For example, if I look at the SBBI Classic Yearbook--a standard reference work and the source of an awful lot of quoted numbers--for 2010, p. 33, table 2-4, "Portfolios," this is what they say; I'm quoting their words and using their definitions. "Large stocks" means S&P 500 and its 90-stock predecessor. "Bonds" means long-term government bonds. Returns are nominal. For a 20-year holding period, the longest they show, out of 65 overlapping 20-year periods. First number is the number of times it was positive, second is the number of time it was the highest returning portfolio:

100% Large Stocks: 65, 58
90% Large Stocks/10% Bonds: 65, 0
70% Large Stocks/30% Bonds: 65, 3
50% Large Stocks/50% Bonds: 65, 3
30% Large Stocks/70% Bonds: 65, 0
10% Large Stocks/90% Bonds: 65, 0
100% Long-Term Government Bonds 65, 0

So, broadly, none of these portfolios ever lost money, 100% stocks was best 58 times out of 65 = 89% of the time, and 50/50 was best only 5/65 = 4.6% of the time.

We all know this. We get it. The question isn't what the data shows, the question is how to act on it.

And I would ask you this: since the data shows that 100% never showed a loss over any 20-year period, so why do you stop at 100%? Why not use 200% leverage?

There are two issues here. One is size of consequences. People who advocate 100% stocks generally seem to emphasize counts. It's an athletics, competition model. You get the gold medal by beating someone else, it doesn't matter by how much. But in investing, counts don't matter. Bill Miller beat the S&P 500 15 years in a row, then underperformed 3 years in a row. He was lionized during the first 15 years of the streak, now he isn't. Why not? He still won 15 times out of 18, didn't he? The Vanguard 500 Index Fund lost to the S&P 500 18 years in a row, Legg Mason Value Trust won 15 times out of 18, which would you rather have owned?

The second issue concerns a) personal risk tolerance, and b) philosophy with regard to rare high-consequences events. What bothers me about your analysis is this. I may be wrong but this is what I think you are doing. You want to be at 100% stocks because of greed, i.e. you think you will make more money that way. Now, you could justify this by saying "I am willing to take the risk," i.e. accept a real possibility of personal financial ruin in exchange for a likelihood of making more. That would be fine. If you really take the risk, you deserve the risk premium.

But I think you are saying something more like this: "As a reward for having the cojones and for staying invested for at least 20 years, the risk actually goes away, and the proof is that in the United States, during the historical record over the period of time during which accurate data was available, the risk never showed up."

Or maybe this: "It's just a roller-coaster ride. It feels dangerous but it's really perfectly safe, for anyone with the strength of character to stay in their seats. It's only dangerous for the scaredycats who jump out, and I'm not one of them."
Annual income twenty pounds, annual expenditure nineteen nineteen and six, result happiness; Annual income twenty pounds, annual expenditure twenty pounds ought and six, result misery.
staythecourse
Posts: 6993
Joined: Mon Jan 03, 2011 8:40 am

Re: Argument for a 100% stock allocation for 20 yr horizon

Post by staythecourse »

I don't need stats to prove it to me. A 100% stock portfolio (diversified geographically) SHOULD have a higher expected return then mix with bonds, otherwise the foundation of investing has a SERIOUS crack in it. One should expected a higher EXPECTED return with they take on a higher ex ante risk, in this case being higher short term volatility.

I don't invest that way, but accept a 100% equity portfolio is the most logical approach for those who: 1. Have a 20+ year horizon, 2. Have enough liquidity in EF not to have to touch the stocks during bumps in one's life, 3. Whose single goal is to produce the most money form their investments at the end, i.e. retirement, 4. Someone who can stay the course over their time horizon, 5. Preferably, one who adds new money in every month to take adv. of the higher volatility, and 6. The person should be prepared to lose up to 50% in any one year being 100% stocks.

The simple reason folks don't do 100% equities is they can't look into to a mirror and say yes to all of the above.

Also, my question to the OP is you have to ask yourself why did you start researching this data now? Don't think it is a coincidince it is when stocks are at a all time high. That is what is called recency bias. I don't remember too many 100% equity folks during 2008 or even in August 2011 or Oct. 2011.

My advice for the folks who do meet my criteria above is ONLY implement your 100% equities after at least a 10-20% drop in the market. This will prove you actually believe in it or you are just market timing.

Good luck.
"The stock market [fluctuation], therefore, is noise. A giant distraction from the business of investing.” | -Jack Bogle
riskreward
Posts: 130
Joined: Sat Jan 05, 2008 10:32 am

Re: Argument for a 100% stock allocation for 20 yr horizon

Post by riskreward »

Fundtalker,

Just plug your assumptions into a Monte Carlo simulation. It contains thousands of historical scenarios.

Use a 20 year period and your withdrawal rate and see what it gives you using differing equity allocations.

I have never seen in my situation where 100% equities has the highest success rate.
User avatar
Don Christy
Posts: 391
Joined: Sun Oct 11, 2009 10:33 pm

Re: Argument for a 100% stock allocation for 20 yr horizon

Post by Don Christy »

The best argument I've heard for 100% equities is that it will increase your vocabulary. You'll sooner or later learn words like capitulate.

:beer

Seriously, good luck.
“Speak only if it improves upon the silence." Mahatma Gandhi
Harold
Posts: 3154
Joined: Sat Mar 03, 2007 6:50 pm
Location: San Francisco

Re: Argument for a 100% stock allocation for 20 yr horizon

Post by Harold »

staythecourse wrote:I don't need stats to prove it to me. A 100% stock portfolio (diversified geographically) SHOULD have a higher expected return then mix with bonds, otherwise the foundation of investing has a SERIOUS crack in it. One should expected a higher EXPECTED return with they take on a higher ex ante risk, in this case being higher short term volatility.
Whether you call it a crack or not, the foundation is less solid than ordinary investors might think. It's more like an investor is planning on the equity risk premium persisting favorably for their investing lifetime, which it very well might given that it has for long periods in U.S. market history. And looked at abstractly, the equity risk premium is reflecting competitive pricing of what could be exceedingly valuable cash flows -- so it makes sense to be able to partake of significant upside (though the pricing can and does often get adjusted downward, sometimes violently).

But at the core, we should remember that at original stock issue (and even now), stocks and bonds are the same value. That is, at the original offering, a company issuing $100M in stocks and $100M in bonds wasn't able to get more than $100M for the stocks (i.e. some of that "obvious" extra value of stocks) -- and even now, the value of the S&P 20 years from now is still selling at the same price as treasuries (i.e. $100M in the S&P 500 index is the same price as $100M in 10-year treasuries). If it were so obvious that for any given time period we can depend on stocks outperforming, prices would reflect that.
555
Posts: 4955
Joined: Thu Dec 24, 2009 6:21 am

Re: Argument for a 100% stock allocation for 20 yr horizon

Post by 555 »

Many target retirement funds are 90% stock until age 40. I think this is reasonable, as is the overall glide path of these funds. Being 100% stock until age 40 is not too crazy, especially if you are making regular contributions (and of course you're not cashing out at age 40) but you do need to move to more bonds as time goes on. I'm talking about the initial 20 years of a much longer investment horizon.

Also, since returns compound, (i.e. you multiply) then it is natural to use geometric mean (not arithmetic) for expected balance. Using this, if you simply wanted to maximize(geometric) expected return, then unlimited leverage is not optimal. I think optimal is about 130% stock, -30% bond, but I may have oversimplified that strategy.
yobria
Posts: 5978
Joined: Mon Feb 19, 2007 10:58 pm
Location: SF CA USA

Re: Argument for a 100% stock allocation for 20 yr horizon

Post by yobria »

100% stocks is fine (its been my allocation for 20 years), but if you go this route you need a Plan B, in case the risk shows up. Plan Bs include:

-Enough human capital to recover (young enough)
-Enough assets to live even on a fraction of expected returns (ability to take risk)
-Some other ace in the hole - inheritence, real estate, pension, etc.

My mother is 62 and has none of the above, so we keep her at 55/45.
The Wizard
Posts: 13356
Joined: Tue Mar 23, 2010 1:45 pm
Location: Reading, MA

Re: Argument for a 100% stock allocation for 20 yr horizon

Post by The Wizard »

We've been over this before in other threads.
100% stocks at a young age is non-optimal because it limits your ability to BUY LOW during stock market crashes.
Young investors have the "advantage" that their accumulation is modest compared to their annual contributions of new money, so that helps. I would aim for a target AA of no more than 90% stocks, maybe even 85% stocks, with the rest in bonds, CDs or similar, depending on the fixed-income "climate".
Then I would adjust my weekly/monthly contributions to maintain your target allocation as best you can. When stocks have their next precipitous drop, new contribs won't be enough and you'll have to move some of those bonds into stocks.
This will be good TRAINING for later on as well...
Attempted new signature...
User avatar
Topic Author
fundtalker123
Posts: 901
Joined: Tue Feb 27, 2007 3:18 am

Re: Argument for a 100% stock allocation for 20 yr horizon

Post by fundtalker123 »

riskreward wrote:Fundtalker,

Just plug your assumptions into a Monte Carlo simulation. It contains thousands of historical scenarios.

Use a 20 year period and your withdrawal rate and see what it gives you using differing equity allocations.

I have never seen in my situation where 100% equities has the highest success rate.
How do you define success rate?
Last edited by fundtalker123 on Sat Mar 17, 2012 3:27 pm, edited 1 time in total.
User avatar
Topic Author
fundtalker123
Posts: 901
Joined: Tue Feb 27, 2007 3:18 am

Re: Argument for a 100% stock allocation for 20 yr horizon

Post by fundtalker123 »

nisiprius wrote: There are two issues here. One is size of consequences. People who advocate 100% stocks generally seem to emphasize counts. It's an athletics, competition model. You get the gold medal by beating someone else, it doesn't matter by how much. But in investing, counts don't matter. Bill Miller beat the S&P 500 15 years in a row, then underperformed 3 years in a row. He was lionized during the first 15 years of the streak, now he isn't. Why not? He still won 15 times out of 18, didn't he? The Vanguard 500 Index Fund lost to the S&P 500 18 years in a row, Legg Mason Value Trust won 15 times out of 18, which would you rather have owned?
My analysis didn't just emphasize counts, it emphasized degree of underperformance or overperformance. The few times 100% stocks did worse, they were worse by only a small amount <13%.
User avatar
Topic Author
fundtalker123
Posts: 901
Joined: Tue Feb 27, 2007 3:18 am

Re: Argument for a 100% stock allocation for 20 yr horizon

Post by fundtalker123 »

mwm158 wrote:
fundtalker123 wrote:
mwm158 wrote:In my several years and several million hands of poker, I went to a showdown and lost to a royal flush exactly twice. Both times I had a boat. The evil part about it is they happened on 2 consecutive nights. The odds of that happening were as close to zero as you can get. I had a million hands worth of historical data telling me this wouldn't happen once, let alone twice. How many sigmas is that? Anyway, my point is that 100 years worth of stock data means absolutely diddly nothing. I wouldn't be 100% in anything.
So your theory for asset allocation is, I don't know anything about what might happen for asset class A or asset class B, so I'll just choose 50/50?
Not quite, my "theory" is based on the idea that if at any point in time equities went to zero, I won't be stuck eating dog food. I continually get more and more conservative as I approach retirement.
Fair enough, but how do you know bonds also couldn't have an absolutely terrible return over 20 years, such that in real return they essentially go to zero.
User avatar
Topic Author
fundtalker123
Posts: 901
Joined: Tue Feb 27, 2007 3:18 am

Re: Argument for a 100% stock allocation for 20 yr horizon

Post by fundtalker123 »

Kevin M wrote:fundtalker123: What is your AA? Has this analysis caused you to reconsider your AA?

Kevin
My asset allocation is 70/30. I have thought about moving towards 50/50 over the next 20 years (currently age 40), but my analysis has caused me to feel like I'd rather stay at 70/30 than glide to 50/50. Occasionally I consider upping stocks further.
mwm158
Posts: 475
Joined: Sat Jun 25, 2011 6:55 pm

Re: Argument for a 100% stock allocation for 20 yr horizon

Post by mwm158 »

deleted
Last edited by mwm158 on Wed Jan 07, 2015 8:05 pm, edited 1 time in total.
HornedToad
Posts: 1130
Joined: Wed May 21, 2008 12:36 am

Re: Argument for a 100% stock allocation for 20 yr horizon

Post by HornedToad »

If you want 100% stocks then go 100% stocks. You don't need to try to justify it or convince other people that it's great and they should also be 100% stocks. Just have confidence in your decision and go for it and see what happens.

That said, I'm 100% stocks (age 29) and have no issues with it and think for my situation it makes sense.
Mortgasm
Posts: 285
Joined: Mon Feb 06, 2012 6:54 pm

Re: Argument for a 100% stock allocation for 20 yr horizon

Post by Mortgasm »

HornedToad wrote:If you want 100% stocks then go 100% stocks. You don't need to try to justify it or convince other people that it's great and they should also be 100% stocks. Just have confidence in your decision and go for it and see what happens.

That said, I'm 100% stocks (age 29) and have no issues with it and think for my situation it makes sense.

Exactly.

But I don't think that's the point the OP is trying to make. I think the he believes he's found an irrefutable position for 100% stocks over 20 years.
rai
Posts: 1342
Joined: Tue Apr 06, 2010 7:11 am

Re: Argument for a 100% stock allocation for 20 yr horizon

Post by rai »

Mortgasm wrote:100% is fine at year 20, sure. But it's not at year 5 because you want to preserve your wins.
yes

so you have $200K you don't mind if it goes down $100K

So you have $5M you WILL mind if it goes down $2.5M

the way I see it is, you don't want bonds until you don't have any and you needed them and it's too late.
Last edited by rai on Sat Mar 17, 2012 9:32 pm, edited 1 time in total.
"Life is what happens to you while you're busy making other plans" - John Lennon. | | "You say that money, isn't everything | But I'd like to see you live without it." - Silverchair
User avatar
GregLee
Posts: 1748
Joined: Wed Oct 27, 2010 3:54 pm
Location: Waimanalo, HI

Re: Argument for a 100% stock allocation for 20 yr horizon

Post by GregLee »

nisiprius wrote:But in investing, counts don't matter.
They do matter. We talked about this issue in another thread. It's better to make more money in investing because it makes you less vulnerable to losses. The more you have, the more you can afford to lose. I'm not saying that only potential gains count, I'm just saying that they do count for something. You seem to discount the extra safety that gains in stocks can confer, and this makes no sense to me.
Greg, retired 8/10.
User avatar
Topic Author
fundtalker123
Posts: 901
Joined: Tue Feb 27, 2007 3:18 am

Re: Argument for a 100% stock allocation for 20 yr horizon

Post by fundtalker123 »

nisiprius wrote:fundtalker, your analysis isn't new or unknown to Bogleheads.
Show me another post on this forum that calculated exactly what I calculated. I think many people here don't know the such details of past history, such as what the frequency of occurrences of certain total returns over a 20 year period were, and because of this some people are possibly being scare-mongered into holding a very high allocation to bonds, and thus risking the chance that inflation and potential declines in interest rate will rapidly erode their savings.
Last edited by fundtalker123 on Sat Mar 17, 2012 3:13 pm, edited 1 time in total.
User avatar
Topic Author
fundtalker123
Posts: 901
Joined: Tue Feb 27, 2007 3:18 am

Re: Argument for a 100% stock allocation for 20 yr horizon

Post by fundtalker123 »

The Wizard wrote:We've been over this before in other threads.
100% stocks at a young age is non-optimal because it limits your ability to BUY LOW during stock market crashes.
If you don't need the money until 20 years from now the only thing that matters is what your total real return at the end was. My calculation showed that in the past almost always it was higher for 100% stocks, and in just 8% of starting years since 1928 it was just a small amount lower, and in the one worst case (1 out of 64 starting years) it was just 13% lower. I don't think everybody knows these numbers, but if you invest in stocks and bonds I think you should know these numbers. That doesn't mean you should necessarily choose 100/0 over 50/50, but it should influence your thinking. I choose 70/30. The future may be worse for stocks, of course.
User avatar
Topic Author
fundtalker123
Posts: 901
Joined: Tue Feb 27, 2007 3:18 am

Re: Argument for a 100% stock allocation for 20 yr horizon

Post by fundtalker123 »

Mortgasm wrote:
HornedToad wrote:If you want 100% stocks then go 100% stocks. You don't need to try to justify it or convince other people that it's great and they should also be 100% stocks. Just have confidence in your decision and go for it and see what happens.

That said, I'm 100% stocks (age 29) and have no issues with it and think for my situation it makes sense.

Exactly.

But I don't think that's the point the OP is trying to make. I think the he believes he's found an irrefutable position for 100% stocks over 20 years.
No, I'm saying I found a nearly irrefutable position for having held 100% stocks for every 20 year period since 1928. The future may not repeat the past, but should influence one's thinking.
User avatar
Topic Author
fundtalker123
Posts: 901
Joined: Tue Feb 27, 2007 3:18 am

Re: Argument for a 100% stock allocation for 20 yr horizon

Post by fundtalker123 »

TheEternalVortex wrote:
fundtalker123 wrote:
TheEternalVortex wrote:There are less then 5 non-overlapping 20 year periods in your sample. How confident can you be based on that?
How confident can I be about what? When you invest, if you do not base your asset allocation on past history of returns of bonds vs. stocks, what do you base it on?
Do you realize the error in concluding that "8%" of the time the 50/50 wins when you only actually have 4.5 such periods? How confident can you be that 100% is a good bet based on 4.5 samples? You don't have 60 different samples because they are all highly correlated with each other.

For example, I randomly generated a list of 60 coin flips (0 or 1). It so happens that in this set, over 92% of sets of 20 consecutive coinflips have more 1s than 0s. Can we conclude anything from this?

1,0,1,1,0,0,1,0,1,0,0,1,1,0,1,1,0,0,1,1,1,0,0,1,1,1,1,1,1,1,0,1,0,0,1,0,0,0,0,1,1,1,1,0,0,1,1,1,0,1,1,0,1,1,0,0,0,0,0,0

(Overall there are only slightly more 1s than 0s: 32 vs 28)
First, stocks doing better than bonds is not a coin flip. There are also theoretical reasons for it. Second, how confident can you be, using exactly the same data, that 50/50, if that's what you prefer, is safer than 100/0 ?
User avatar
Topic Author
fundtalker123
Posts: 901
Joined: Tue Feb 27, 2007 3:18 am

Re: Argument for a 100% stock allocation for 20 yr horizon

Post by fundtalker123 »

nisiprius wrote:No, because I'm not interested in getting the highest possible average total return--if I were, I wouldn't stop at 100% stocks, I'd use leverage.

Leverage isn't free, however.
User avatar
Topic Author
fundtalker123
Posts: 901
Joined: Tue Feb 27, 2007 3:18 am

Re: Argument for a 100% stock allocation for 20 yr horizon

Post by fundtalker123 »

mwm158 wrote:
fundtalker123 wrote: Fair enough, but how do you know bonds also couldn't have an absolutely terrible return over 20 years, such that in real return they essentially go to zero.
If bad things happen, I can ... become a male prostitute, etc.
Ok
User avatar
momar
Posts: 1359
Joined: Sun Nov 13, 2011 11:51 am

Re: Argument for a 100% stock allocation for 20 yr horizon

Post by momar »

fundtalker123 wrote:
The Wizard wrote:We've been over this before in other threads.
100% stocks at a young age is non-optimal because it limits your ability to BUY LOW during stock market crashes.
If you don't need the money until 20 years from now the only thing that matters is what your total real return at the end was. My calculation showed that in the past almost always it was higher for 100% stocks, and in just 8% of starting years since 1928 it was just a small amount lower, and in the one worst case (1 out of 64 starting years) it was just 13% lower. I don't think everybody knows these numbers, but if you invest in stocks and bonds I think you should know these numbers. That doesn't mean you should necessarily choose 100/0 over 50/50, but it should influence your thinking. I choose 70/30. The future may be worse for stocks, of course.
You are missing the point that once you have held for, say, 10 years, the next 10 years does not care what has already happened. Your time horizon is now effectively 10 years. Then effectively 9, 8, 7, 6 and so on.

Essentially, at each instant in time you have the ability to make a decision as to what the best AA is going forward. And at each instant, your circumstances change. How does holding 100/0 for 20 years compare to holding 100/0 for 15 years and then 50/50 for the remaining 5? How about for every possible scenario?

You have only performed the most basic of backtesting and have barely scratched the surface.
"Index funds have a place in your portfolio, but you'll never beat the index with them." - Words of wisdom from a Fidelity rep
tibbitts
Posts: 23728
Joined: Tue Feb 27, 2007 5:50 pm

Re: Argument for a 100% stock allocation for 20 yr horizon

Post by tibbitts »

Unless you're investing for something other than your personal future, a significant problem with 100% stocks is that you can never count on any particular timeline. You can't assume you'll not need or want as much of your money as possible next week, much less in 10 or 20 years.

Paul
Mortgasm
Posts: 285
Joined: Mon Feb 06, 2012 6:54 pm

Re: Argument for a 100% stock allocation for 20 yr horizon

Post by Mortgasm »

The troll is strong in this one.
User avatar
Topic Author
fundtalker123
Posts: 901
Joined: Tue Feb 27, 2007 3:18 am

Re: Argument for a 100% stock allocation for 20 yr horizon

Post by fundtalker123 »

tibbitts wrote:Unless you're investing for something other than your personal future, a significant problem with 100% stocks is that you can never count on any particular timeline. You can't assume you'll not need or want as much of your money as possible next week, much less in 10 or 20 years.

Paul
To say you can't assume you'll not need or want as much of your money as possible next week is a bit silly. I don't need my savings next week, which would also involve paying a penalty to withdraw from retirement accounts, I just need the part of my paycheck not going to savings.
tibbitts
Posts: 23728
Joined: Tue Feb 27, 2007 5:50 pm

Re: Argument for a 100% stock allocation for 20 yr horizon

Post by tibbitts »

fundtalker123 wrote:
tibbitts wrote:Unless you're investing for something other than your personal future, a significant problem with 100% stocks is that you can never count on any particular timeline. You can't assume you'll not need or want as much of your money as possible next week, much less in 10 or 20 years.

Paul
To say you can't assume you'll not need or want as much of your money as possible next week is a bit silly. I don't need my savings next week, which would also involve paying a penalty to withdraw from retirement accounts, I just need the part of my paycheck not going to savings.
You might change your mind if you find out that you or someone in your family needs an expensive experimental medical procedure that your insurance won't pay for, or simply realize that you have only some months of good health left and want to enjoy your money (rather than 50% or less of it, in the event of an untimely downturn) while you're alive. Obviously, this would only apply to someone who, like most of us, doesn't have so large an amount of savings that it would be almost impossible to spend a large portion of it in a short period of time.

Paul
555
Posts: 4955
Joined: Thu Dec 24, 2009 6:21 am

Re: Argument for a 100% stock allocation for 20 yr horizon

Post by 555 »

Mortgasm wrote: I think the he believes he's found an irrefutable position for 100% stocks over 20 years.
This statement is not true.
Mortgasm
Posts: 285
Joined: Mon Feb 06, 2012 6:54 pm

Re: Argument for a 100% stock allocation for 20 yr horizon

Post by Mortgasm »

555 wrote:
Mortgasm wrote: I think the he believes he's found an irrefutable position for 100% stocks over 20 years.
This statement is not true.
That I think it? Or that he believes he found it? Or that he did find it?

I definitely think he believes that.
Post Reply