The Efficient Frontier of Stocks and Bonds

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
TheCluelessInvestor
Posts: 46
Joined: Thu Feb 09, 2012 8:12 pm
Location: Cincinnati, OH

The Efficient Frontier of Stocks and Bonds

Post by TheCluelessInvestor » Wed Apr 11, 2012 2:37 pm

Looking for your opinions on this:

I was curious about what an efficient frontier graph of stocks and bonds would look like over different time periods, so I created the graphs below using the Russell 3000 and Barclays Aggregate indices. The portfolios were rebalanced every year.

Image

Image

The first graph demonstrates what different allocations of stocks and bonds (10% increments) would look like for a 10, 20, or 30 year period ending 03/31/2012. It shows that the longer the time period the better a 100% stock allocation does. While there may be a benefit of adding up to 20% in bonds to the portfolio as far as risk goes, there was no benefit to overall return, the exception being the 10 year line, which shows that, generally, the more bonds a portfolio held, the better it did. I figured that the reason was the crazy markets we had over the past decade, especially since the housing bubble. I decided to run a similar graph, but this time for periods ending 03/31/2012. The second graph shows that while adding bonds to the stock portfolio do reduce risk, it also comes at a high cost due to the steep decline in returns (I used 10, 15, and 20 year time periods since I only had returns going back to 1979).

I also wondered about the “your age in bonds” rule. So I created a portfolio for someone retiring in 2012 at age 65. So in 2012, that person would have 65% in bonds and 35% in stocks. In 2011, it would have been 64% in bonds and 36% in stocks. You get the idea. I plotted those figures in the first graph under YAIB (XX Yr). It shows that a simple portfolio holding a constant allocation to bonds between 50% and 60% would have achieved close to the same results over the different time periods without the need to rebalance it based on the person’s age.

I concluded that someone who started saving 20 or 30 years ago without allocating any money to bonds would still be able to retire with more money, even after the most recent market crash, than someone with a more conservative approach. I don’t think I would be 100% in stock when close to retirement myself, but still found this interesting.

What do you guys think?

boglestan
Posts: 193
Joined: Wed Mar 30, 2011 1:52 pm

Re: The Efficient Frontier of Stocks and Bonds

Post by boglestan » Wed Apr 11, 2012 2:57 pm

I've seen other analyses with similar results. I think it shows that the age in bonds rule is mostly a psychological tactic. Sure, you could achieve nearly the same result with a constant 50/50 portfolio, or even do a bit better with 100% stocks, but I think for most people, it would be extremely difficult to be healthy at age 75 and see your portfolio drop by 30% or 50%, even if you know that it will most likely recover. It might be gut wrenching. Who wants a gut wrenching retirement? That's what age in bonds is for.

See this reddit post for more: http://www.reddit.com/r/portfolios/comm ... e_insight/

staythecourse
Posts: 5075
Joined: Mon Jan 03, 2011 9:40 am

Re: The Efficient Frontier of Stocks and Bonds

Post by staythecourse » Wed Apr 11, 2012 7:53 pm

I've always thought the age in bonds was rather simplified for the masses.

My advice for those who want to be aggressive: 80/20 from the beginning until 10 years before you need the money. Every rolling 10 yr. period in history with a 80/20 was positive in nominal terms (the way folks look at their money value). This would give more then enough time to recoup any losses on such a high equity allocation before it would be needed to be spent.

Good luck.
"The stock market [fluctuation], therefore, is noise. A giant distraction from the business of investing.” | -Jack Bogle

Prudence
Posts: 277
Joined: Fri Mar 09, 2012 4:55 pm

Re: The Efficient Frontier of Stocks and Bonds

Post by Prudence » Thu Apr 12, 2012 9:15 am

staythecourse wrote:I've always thought the age in bonds was rather simplified for the masses.

My advice for those who want to be aggressive: 80/20 from the beginning until 10 years before you need the money. Every rolling 10 yr. period in history with a 80/20 was positive in nominal terms (the way folks look at their money value). This would give more then enough time to recoup any losses on such a high equity allocation before it would be needed to be spent.

Good luck.


Staythecourse, what is your strategy, instead of age in bonds, for those who hit 65 or normal retirement age and must plan to be around for 30 years (looking for broad generalization here)?

TheCluelessInvestor
Posts: 46
Joined: Thu Feb 09, 2012 8:12 pm
Location: Cincinnati, OH

Re: The Efficient Frontier of Stocks and Bonds

Post by TheCluelessInvestor » Thu Apr 12, 2012 9:24 am

Prudence,

What is your goal during retirement? How much money do you have? Are you short and need to be aggressive? Do you have enough and all you care about is income? Or maybe you have more than enough and are planning on leaving your money to your children, in which case you should invest according to their age, not yours? There are so many variables; it is hard to get a short/right answer on a blog. But don’t listen to me, I’m clueless! :wink:

Prudence
Posts: 277
Joined: Fri Mar 09, 2012 4:55 pm

Re: The Efficient Frontier of Stocks and Bonds

Post by Prudence » Thu Apr 12, 2012 9:31 am

I would like to earn about inflation plus two percent over the long term.

User avatar
Lbill
Posts: 4997
Joined: Thu Mar 13, 2008 11:25 pm
Location: Somewhere between Up and Down

Re: The Efficient Frontier of Stocks and Bonds

Post by Lbill » Thu Apr 12, 2012 9:47 am

I also wondered about the “your age in bonds” rule. So I created a portfolio for someone retiring in 2012 at age 65. So in 2012, that person would have 65% in bonds and 35% in stocks. In 2011, it would have been 64% in bonds and 36% in stocks. You get the idea. I plotted those figures in the first graph under YAIB (XX Yr). It shows that a simple portfolio holding a constant allocation to bonds between 50% and 60% would have achieved close to the same results over the different time periods without the need to rebalance it based on the person’s age.

Thanks for this analysis. I also performed an analysis to compare the returns of AIB to a fixed allocation with a portfolio withdrawal rate of 4% real. As others have found, the fixed allocation lasts longer in decumulation than the AIB allocation (ie. produces higher returns). The fixed allocation I looked at was maintaining the average of the AIB allocation for a 65-year old retiree (which is 82.5%) throughout. One of the reasons this works is that it has lower sequence-of-returns risk, which the risk that stocks do the worst in the early years of retirement when your equity allocation is highest using AIB. Since the statistical evidence shows that AIB doesn't actually improve returns, I conclude the reason it is still advocated is primarily psychological.
"Life can only be understood backward; but it must be lived forward." ~ Søren Kierkegaard | | "You can't connect the dots looking forward; but only by looking backwards." ~ Steve Jobs

555
Posts: 4955
Joined: Thu Dec 24, 2009 7:21 am

Re: The Efficient Frontier of Stocks and Bonds

Post by 555 » Thu Apr 12, 2012 11:00 am

Okay you have 6 periods, and stocks beat bonds 5 times out of 6. Isn't this pretty much what you'd expect (depending how you choose the period).

Also the `conclusion' (drawn how?) that while accumulating some constant allocation, say X% stocks, gets the same ultimate result as some glide path. So what? How do you choose X% other than by hindsight. It's non-actionable. An it's virtually a tautology that some such number X% exists. Big whoop!

Also constant allocation gets riskier as time goes on due to decreasing human capital.

TheCluelessInvestor
Posts: 46
Joined: Thu Feb 09, 2012 8:12 pm
Location: Cincinnati, OH

Re: The Efficient Frontier of Stocks and Bonds

Post by TheCluelessInvestor » Thu Apr 12, 2012 11:04 am

555 wrote:Okay you have 6 periods, and stocks beat bonds 5 times out of 6. Isn't this pretty much what you'd expect (depending how you choose the period).

Also the `conclusion' (drawn how?) that while accumulating some constant allocation, say X% stocks, gets the same ultimate result as some glide path. So what? How do you choose X% other than by hindsight. It's non-actionable. An it's virtually a tautology that some such number X% exists. Big whoop!

Also constant allocation gets riskier as time goes on due to decreasing human capital.


Such hostility!

How would you do it?

ourbrooks
Posts: 1575
Joined: Fri Nov 13, 2009 4:56 pm

Re: The Efficient Frontier of Stocks and Bonds

Post by ourbrooks » Thu Apr 12, 2012 11:36 am

It's not at all true that a constant allocation gets riskier over time because of human capital reduction. At the same time as human capital is declining, they've also likely to have acquired lots of income which is non-volatile (riskless) in the form of Social Security, pensions and annuities. Risk reduction comes "for free" without any need to change asset allocation.

What people need to really worry about is not stock market gyrations but the risk of outliving their money. The safe withdrawal rate studies mostly use 30 years periods. This is too low the moment you begin to consider the finances of a couple. For a woman 65 years old married to a man 58 years old, the median dual life expectancy is 30 years; half of the couples with those ages will live longer than the 30 years of the safe withdrawal rate studies. For many couples, the real planning horizon ought to be 40 years, not 30; I'd really love to see the frontier curves extended out over 40 years.

555
Posts: 4955
Joined: Thu Dec 24, 2009 7:21 am

Re: The Efficient Frontier of Stocks and Bonds

Post by 555 » Thu Apr 12, 2012 11:56 am

ourbrooks wrote:It's not at all true that a constant allocation gets riskier over time because of human capital reduction. At the same time as human capital is declining, they've also likely to have acquired lots of income which is non-volatile (riskless) in the form of Social Security, pensions and annuities. Risk reduction comes "for free" without any need to change asset allocation.

What people need to really worry about is not stock market gyrations but the risk of outliving their money. The safe withdrawal rate studies mostly use 30 years periods. This is too low the moment you begin to consider the finances of a couple. For a woman 65 years old married to a man 58 years old, the median dual life expectancy is 30 years; half of the couples with those ages will live longer than the 30 years of the safe withdrawal rate studies. For many couples, the real planning horizon ought to be 40 years, not 30; I'd really love to see the frontier curves extended out over 40 years.

Personally I have no Social Security, pensions or annuities (I'd buy an annuity sometime during retirement). To the extent that someone has those, it's true that they increase while human capital decreases, but human capital is a bigger consideration, but in any case those things just mean you can be overall riskier than if you didn't have them, but it doesn't argue for going constant instead of glide path.

Of course if you are accumulating more than enough, you have much more flexibility in plan.

Rodc
Posts: 13601
Joined: Tue Jun 26, 2007 9:46 am

Re: The Efficient Frontier of Stocks and Bonds

Post by Rodc » Thu Apr 12, 2012 11:58 am

This looks like you used one time period?

For example 10 years is 2003-2012?

20 years is 1993-2012?

30 years is 1983-2012?

Or is this rolling periods?
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.

TheCluelessInvestor
Posts: 46
Joined: Thu Feb 09, 2012 8:12 pm
Location: Cincinnati, OH

Re: The Efficient Frontier of Stocks and Bonds

Post by TheCluelessInvestor » Thu Apr 12, 2012 12:00 pm

This:

Rodc wrote:For example 10 years is 2003-2012?

20 years is 1993-2012?

30 years is 1983-2012?

Rodc
Posts: 13601
Joined: Tue Jun 26, 2007 9:46 am

Re: The Efficient Frontier of Stocks and Bonds

Post by Rodc » Thu Apr 12, 2012 12:07 pm

TheCluelessInvestor wrote:This:

Rodc wrote:For example 10 years is 2003-2012?

20 years is 1993-2012?

30 years is 1983-2012?


Thanks.

I would suggest the results are not very meaningful.

In the 30 year you have a huge 20 year bull that is entirely missing from the 10 year, so very much apples to oranges.

It is unclear that any of these periods are "typical", "average", or "representative". So I don't know that they contain much of a lesson for how to invest or what to expect.

It is a real challenge to do meaningful historical analyses. Even 90 years of data only gives 3 independent 30-year periods. One can't much from 3 data points. Even if you start in 1926 and use overlapping 30-year periods as is commonly done, the crash of 1929 shows up in only 4 points whereas 1950 shows up in 30 points, and the crashes of 2001 and 2008 show up zero times.

Not to say looking at history has no value. History is the only thing we have that is real. Just don't expect history to tell you very much that is solid.
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.

TheCluelessInvestor
Posts: 46
Joined: Thu Feb 09, 2012 8:12 pm
Location: Cincinnati, OH

Re: The Efficient Frontier of Stocks and Bonds

Post by TheCluelessInvestor » Thu Apr 12, 2012 12:18 pm

Rodc wrote:
TheCluelessInvestor wrote:This:

Rodc wrote:For example 10 years is 2003-2012?

20 years is 1993-2012?

30 years is 1983-2012?


Thanks.

I would suggest the results are not very meaningful.

In the 30 year you have a huge 20 year bull that is entirely missing from the 10 year, so very much apples to oranges.

It is unclear that any of these periods are "typical", "average", or "representative". So I don't know that they contain much of a lesson for how to invest or what to expect.

It is a real challenge to do meaningful historical analyses. Even 90 years of data only gives 3 independent 30-year periods. One can't much from 3 data points. Even if you start in 1926 and use overlapping 30-year periods as is commonly done, the crash of 1929 shows up in only 4 points whereas 1950 shows up in 30 points, and the crashes of 2001 and 2008 show up zero times.

Not to say looking at history has no value. History is the only thing we have that is real. Just don't expect history to tell you very much that is solid.


The whole point was to illustrate how over the long term, even a 100% allocation to stocks MAY be better, it certainly was over past periods, than a blend between stocks and bonds. Those periods include the crazy market of the past 10 years. Even if someone was invested fully in stocks, as long as he or she had been for a while, that 100% allocation would still have done better, even over the past few years, than any allocation to bonds at all. If you eliminate the last 10 years, it does even better. If someone has a higher risk tolerance (young, no short term need for cash), investing any portion at all in bonds, may not do anything at all. I personally don't have any bonds. If the markets tank, I will just sit around and wait. I know in the end I am most likely to do better than trying to time it.

ourbrooks
Posts: 1575
Joined: Fri Nov 13, 2009 4:56 pm

Re: The Efficient Frontier of Stocks and Bonds

Post by ourbrooks » Thu Apr 12, 2012 12:48 pm

555 wrote:Personally I have no Social Security, pensions or annuities (I'd buy an annuity sometime during retirement). To the extent that someone has those, it's true that they increase while human capital decreases, but human capital is a bigger consideration, but in any case those things just mean you can be overall riskier than if you didn't have them, but it doesn't argue for going constant instead of glide path.

Of course if you are accumulating more than enough, you have much more flexibility in plan.


There's two kinds of risk involved here: risk, in the sense of volatility, how much your portfolio goes up and down from year to year and risk of living too long and running out of money. Glidepaths address only the first kind of risk; as posts in this thread show, they might actually increase the second kind of risk. Suppose that the next 30 years hold some periods of high inflation; would a portfolio high in bonds generate enough earnings to prevent depleting your portfolio prematurely?

The only safe way to play glidepaths is to reduce your withdrawal rate at the same time. Glide down from 4% to 3% or 2%. Of course, only those people who are accumulating more than enough can keep up their living standard while reducing their withdrawal rates. It's the people who don't have excess amounts who might be safer with the constant allocation.

staythecourse
Posts: 5075
Joined: Mon Jan 03, 2011 9:40 am

Re: The Efficient Frontier of Stocks and Bonds

Post by staythecourse » Thu Apr 12, 2012 1:38 pm

Prudence wrote:
staythecourse wrote:I've always thought the age in bonds was rather simplified for the masses.

My advice for those who want to be aggressive: 80/20 from the beginning until 10 years before you need the money. Every rolling 10 yr. period in history with a 80/20 was positive in nominal terms (the way folks look at their money value). This would give more then enough time to recoup any losses on such a high equity allocation before it would be needed to be spent.

Good luck.


Staythecourse, what is your strategy, instead of age in bonds, for those who hit 65 or normal retirement age and must plan to be around for 30 years (looking for broad generalization here)?


Not that I am anywhere close to being there, but I like this approach:

Average your annual income needs the last couple of years before retirement. I then would do 5 years of that amount in a cash (I would include pension and SS as part of the cash). I would then take another 5 years worth and put it into an intermediate term bond fund. The rest I would put into stocks according to your willingness, ability, and need to take risk. The amount in stocks matters at this point is to your risk tolerance, how good a health you and your husband/ wife are in, goals of passing on money to heirs, etc...

I like it because it splits your money into needs in the short term (<10yrs) and long term (>10yrs). It separates the money in the short term protecting against volatility (principal stable investments) and long term from inflation risk (high growth assets). I like the separate bucket list for retirement as you have two competing issues of needs of short term stability of income, but also future need of growth to outpace inflation.

Good luck.
"The stock market [fluctuation], therefore, is noise. A giant distraction from the business of investing.” | -Jack Bogle

Prudence
Posts: 277
Joined: Fri Mar 09, 2012 4:55 pm

Re: The Efficient Frontier of Stocks and Bonds

Post by Prudence » Thu Apr 12, 2012 2:05 pm

Staythecourse, thank you.

555
Posts: 4955
Joined: Thu Dec 24, 2009 7:21 am

Re: The Efficient Frontier of Stocks and Bonds

Post by 555 » Thu Apr 12, 2012 2:17 pm

@ourbrooks, personally I'd glide from risky to more conservative while accumulating then eventially reach a constant allocation during retirement. Basically I'm copying the
Vanguard Target Retirement Funds
https://personal.vanguard.com/us/funds/ ... rementList

@TheCluelessInvestor (OP)
If you want to go 100% stocks that's your choice. It's high risk high expected return. Personally I'm 80:20, and I'm sure not doing age in bonds, but I am gliding more conservative. I want to reach a more conservative allocation in retirement (for the usual reasons), and I think it's better to change gradually than to make a sudden jump. (Sudden moves are risky, and involve guessing timing.)

TheCluelessInvestor
Posts: 46
Joined: Thu Feb 09, 2012 8:12 pm
Location: Cincinnati, OH

Re: The Efficient Frontier of Stocks and Bonds

Post by TheCluelessInvestor » Thu Apr 12, 2012 2:44 pm

@555

I'm not saying I will be 100% in stocks by age 64 myself, but I think people may me missing out by having too much bond exposure, especially at a younger age. The Vanguard fund you referenced to has a 20% bond exposure for people between 45 and 49 years old. Even their 2010 target retirement fund has only a little over 50% in stocks! I'm only 30 years old, so I still have a lot, gosh A LOT of time until retirement. How old are you if you don't mind me asking? So I think any bond exposure I may have now may preclude me from higher returns. Risk is not a big issue since I am not planning on touching that money. Right now I have 60% in VTI and 10% each in VNQ, VB, EPP, and VWO. I think that is a well diversified equity exposure. Based on an efficient frontier graph, it gives me less risk (based on standard deviation) and higher expected returns, over the long run of course. Right now the S&P 500 is up by 1.33% while my portfolio is up by 1.63%. I may get hit big time when the market goes down, like for the past couple weeks, but the opposite is also true, like the past couple days. Since I am investing for the long run, and I believe to be well balanced across broad markets, and that the markets are most likely to be up several years down the road, I'm not too worried about any short term volatility. This will all change of course with time, but I don't think I will be buying bonds anytime soon.

steve r
Posts: 503
Joined: Mon Feb 13, 2012 8:34 pm

Re: The Efficient Frontier of Stocks and Bonds

Post by steve r » Thu Apr 12, 2012 6:21 pm

Your instinct to invest in a broad basket is smarter that what I did years ago when I thought as you do, do nothing but go for the maximum return.
I found draw downs are hard, very hard. I needed the money at the wrong time.

That said ...

With the max return mindset in in mind, I encourage you to look at intermidiate term treasuries as a diversifier - not a total bond fund because that contains corporate bonds which increases its corrleation with stocks). During bull markets, owning stocks is probably preferable to owning corporate bonds. Historically, for over 100 years Treasuries go up when stocks do poorly.

A small amount of intermediate bonds adds to the return and reduces risk To me, one should own bonds for diversification benefits - not returns.

Some baskets - 1972 to 2011 using the Simbia data found on Bogleheads.

TSM 100% return of 9.68
TSM 95% and 5% VFITX - return of 9.69
TSM 90% and 10% VFITX 9.69 ... a higher return and higher risk.
TSM 80% and 20% VFITX 9.65 ... a very similar return with much lower risk.

Try something similar with your 60 /10 / 10 /10 investment and add a treasury bond fund ... maybe 50 / 10 / 10 / 10 /10

That said, even with the chart you show, being 10 percent bonds reduces your performance very little while significantly reducing risk.

Try the backtest and understand the risk / reward tradeoff for your self.
Maximize Diversification - Minimize Costs - Avoid Lotteries

Rodc
Posts: 13601
Joined: Tue Jun 26, 2007 9:46 am

Re: The Efficient Frontier of Stocks and Bonds

Post by Rodc » Thu Apr 12, 2012 7:29 pm

The Vanguard fund you referenced to has a 20% bond exposure for people between 45 and 49 years old. Even their 2010 target retirement fund has only a little over 50% in stocks!


You do understand that large numbers of people, including many professionals and Jack Bogle himself consider that far too many stocks, right?

You recall that many such funds have been slammed in the press for having too much in stocks, right?

Not that you have to agree of course.
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.

TheCluelessInvestor
Posts: 46
Joined: Thu Feb 09, 2012 8:12 pm
Location: Cincinnati, OH

Re: The Efficient Frontier of Stocks and Bonds

Post by TheCluelessInvestor » Thu Apr 12, 2012 8:05 pm

Rodc wrote:
The Vanguard fund you referenced to has a 20% bond exposure for people between 45 and 49 years old. Even their 2010 target retirement fund has only a little over 50% in stocks!


You do understand that large numbers of people, including many professionals and Jack Bogle himself consider that far too many stocks, right?

You recall that many such funds have been slammed in the press for having too much in stocks, right?

Not that you have to agree of course.


Bogle is also against ETFs. Bogle is also against international exposure. Your point is?

If you have more than enough money in retirement, that may be too little equity exposure. If you don't have enough money, that may be too much equity exposure. That fund is a one size fits all.

Rodc
Posts: 13601
Joined: Tue Jun 26, 2007 9:46 am

Re: The Efficient Frontier of Stocks and Bonds

Post by Rodc » Fri Apr 13, 2012 6:53 am

Bogle is also against ETFs. Bogle is also against international exposure. Your point is?


My point is your view, while not unique is a not mainstream.

It is generally recognized that the very high bond allocations in target date fund arose from marketing concerns during a period when stocks were doing well, and did not serve most investors well when stocks tanked. That is why knowledgeable investors recommend picking a target date funds based on desired level of bonds (risk) and not the date associated with the fund.

If you wish to have even higher stock allocations, especially as you get older and farther along in your journey, by all means do so. But I humbly suggest that as you do get farther along, your views may change as you become more knowledgeable. Not a criticism. We all start out as cluelessinvestors.
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.

FinanceFun
Posts: 722
Joined: Mon Nov 28, 2011 9:29 am

Re: The Efficient Frontier of Stocks and Bonds

Post by FinanceFun » Fri Apr 13, 2012 7:02 am

TheCluelessInvestor wrote:@555

I'm not saying I will be 100% in stocks by age 64 myself, but I think people may me missing out by having too much bond exposure, especially at a younger age. The Vanguard fund you referenced to has a 20% bond exposure for people between 45 and 49 years old. Even their 2010 target retirement fund has only a little over 50% in stocks! I'm only 30 years old, so I still have a lot, gosh A LOT of time until retirement. How old are you if you don't mind me asking? So I think any bond exposure I may have now may preclude me from higher returns. Risk is not a big issue since I am not planning on touching that money. Right now I have 60% in VTI and 10% each in VNQ, VB, EPP, and VWO. I think that is a well diversified equity exposure. Based on an efficient frontier graph, it gives me less risk (based on standard deviation) and higher expected returns, over the long run of course. Right now the S&P 500 is up by 1.33% while my portfolio is up by 1.63%. I may get hit big time when the market goes down, like for the past couple weeks, but the opposite is also true, like the past couple days. Since I am investing for the long run, and I believe to be well balanced across broad markets, and that the markets are most likely to be up several years down the road, I'm not too worried about any short term volatility. This will all change of course with time, but I don't think I will be buying bonds anytime soon.



Clueless... Mind if I ask the size of your portfolio? If you expect a 30% loss possibility in equities, it's easy to be cavalier on a $100k portfolio - max loss of $30k. If your max loss is $300k or $600k, that appetite to be cavalier tends to diminish.

rkhusky
Posts: 4545
Joined: Thu Aug 18, 2011 8:09 pm

Re: The Efficient Frontier of Stocks and Bonds

Post by rkhusky » Fri Apr 13, 2012 8:50 am

I am also suspicious of analysis that uses a single start or end date. Averaging over a range of end dates from 1992 - 2012 would make me more comfortable.

TheCluelessInvestor
Posts: 46
Joined: Thu Feb 09, 2012 8:12 pm
Location: Cincinnati, OH

Re: The Efficient Frontier of Stocks and Bonds

Post by TheCluelessInvestor » Fri Apr 13, 2012 9:00 am

steve r wrote:Your instinct to invest in a broad basket is smarter that what I did years ago when I thought as you do, do nothing but go for the maximum return.
I found draw downs are hard, very hard. I needed the money at the wrong time.

That said ...

With the max return mindset in in mind, I encourage you to look at intermidiate term treasuries as a diversifier - not a total bond fund because that contains corporate bonds which increases its corrleation with stocks). During bull markets, owning stocks is probably preferable to owning corporate bonds. Historically, for over 100 years Treasuries go up when stocks do poorly.

A small amount of intermediate bonds adds to the return and reduces risk To me, one should own bonds for diversification benefits - not returns.

Some baskets - 1972 to 2011 using the Simbia data found on Bogleheads.

TSM 100% return of 9.68
TSM 95% and 5% VFITX - return of 9.69
TSM 90% and 10% VFITX 9.69 ... a higher return and higher risk.
TSM 80% and 20% VFITX 9.65 ... a very similar return with much lower risk.

Try something similar with your 60 /10 / 10 /10 investment and add a treasury bond fund ... maybe 50 / 10 / 10 / 10 /10

That said, even with the chart you show, being 10 percent bonds reduces your performance very little while significantly reducing risk.

Try the backtest and understand the risk / reward tradeoff for your self.


Thanks for some CONSTRUCTIVE criticism. I reran the graph using the Russell 3000 and the Barclays Intermediate Treasury indexes and came back with similar results as of 03/31/2012. It would probably be the same as the original as of 03/31/2002:

Image

Over longer time periods, the only thing treasuries seem to do is reduce risk, and also returns. It's up to each individual to figure out what his or her risk tolerance is. From the few things I have tried, it seems that adding a less risky asset to a riskier asset reduces risk and return. But if you add a riskier asset to a less risky one you can achieve both. Look on the other end of the graph. By Adding just 10 or 20% of equities to an all bond portfolio, you can reduce risk AND increase returns.

TheCluelessInvestor
Posts: 46
Joined: Thu Feb 09, 2012 8:12 pm
Location: Cincinnati, OH

Re: The Efficient Frontier of Stocks and Bonds

Post by TheCluelessInvestor » Fri Apr 13, 2012 9:03 am

FinanceFun wrote:
TheCluelessInvestor wrote:@555

I'm not saying I will be 100% in stocks by age 64 myself, but I think people may me missing out by having too much bond exposure, especially at a younger age. The Vanguard fund you referenced to has a 20% bond exposure for people between 45 and 49 years old. Even their 2010 target retirement fund has only a little over 50% in stocks! I'm only 30 years old, so I still have a lot, gosh A LOT of time until retirement. How old are you if you don't mind me asking? So I think any bond exposure I may have now may preclude me from higher returns. Risk is not a big issue since I am not planning on touching that money. Right now I have 60% in VTI and 10% each in VNQ, VB, EPP, and VWO. I think that is a well diversified equity exposure. Based on an efficient frontier graph, it gives me less risk (based on standard deviation) and higher expected returns, over the long run of course. Right now the S&P 500 is up by 1.33% while my portfolio is up by 1.63%. I may get hit big time when the market goes down, like for the past couple weeks, but the opposite is also true, like the past couple days. Since I am investing for the long run, and I believe to be well balanced across broad markets, and that the markets are most likely to be up several years down the road, I'm not too worried about any short term volatility. This will all change of course with time, but I don't think I will be buying bonds anytime soon.



Clueless... Mind if I ask the size of your portfolio? If you expect a 30% loss possibility in equities, it's easy to be cavalier on a $100k portfolio - max loss of $30k. If your max loss is $300k or $600k, that appetite to be cavalier tends to diminish.


It doesn't matter whether it is a $1,000 or $100,000,000 portfolio. If you can't handle the ups and downs of the the market, and you are not in for the long term, you shouldn't be in it at all. If you are preparing yourself for a 30% loss, will you be OK when the markets rise by 30% and you don't get that returns because you were hedging against the loss? It is a two way street. Buy, hold, and forget about it.

TheCluelessInvestor
Posts: 46
Joined: Thu Feb 09, 2012 8:12 pm
Location: Cincinnati, OH

Re: The Efficient Frontier of Stocks and Bonds

Post by TheCluelessInvestor » Fri Apr 13, 2012 9:05 am

rkhusky wrote:I am also suspicious of analysis that uses a single start or end date. Averaging over a range of end dates from 1992 - 2012 would make me more comfortable.


Six different time periods across two different end dates. I don't think of running an analysis that runs for 10, 20, or 30 years as picking specific time periods to make it look better.

FinanceFun
Posts: 722
Joined: Mon Nov 28, 2011 9:29 am

Re: The Efficient Frontier of Stocks and Bonds

Post by FinanceFun » Fri Apr 13, 2012 9:46 am

TheCluelessInvestor wrote:
FinanceFun wrote:
TheCluelessInvestor wrote:@555

I'm not saying I will be 100% in stocks by age 64 myself, but I think people may me missing out by having too much bond exposure, especially at a younger age. The Vanguard fund you referenced to has a 20% bond exposure for people between 45 and 49 years old. Even their 2010 target retirement fund has only a little over 50% in stocks! I'm only 30 years old, so I still have a lot, gosh A LOT of time until retirement. How old are you if you don't mind me asking? So I think any bond exposure I may have now may preclude me from higher returns. Risk is not a big issue since I am not planning on touching that money. Right now I have 60% in VTI and 10% each in VNQ, VB, EPP, and VWO. I think that is a well diversified equity exposure. Based on an efficient frontier graph, it gives me less risk (based on standard deviation) and higher expected returns, over the long run of course. Right now the S&P 500 is up by 1.33% while my portfolio is up by 1.63%. I may get hit big time when the market goes down, like for the past couple weeks, but the opposite is also true, like the past couple days. Since I am investing for the long run, and I believe to be well balanced across broad markets, and that the markets are most likely to be up several years down the road, I'm not too worried about any short term volatility. This will all change of course with time, but I don't think I will be buying bonds anytime soon.



Clueless... Mind if I ask the size of your portfolio? If you expect a 30% loss possibility in equities, it's easy to be cavalier on a $100k portfolio - max loss of $30k. If your max loss is $300k or $600k, that appetite to be cavalier tends to diminish.


It doesn't matter whether it is a $1,000 or $100,000,000 portfolio. If you can't handle the ups and downs of the the market, and you are not in for the long term, you shouldn't be in it at all. If you are preparing yourself for a 30% loss, will you be OK when the markets rise by 30% and you don't get that returns because you were hedging against the loss? It is a two way street. Buy, hold, and forget about it.


That answers my question. Theory is great, till you need to execute. Perhaps after a few years of execution, and accumulating an investment portfolio of some size, you will come to understand.

TheCluelessInvestor
Posts: 46
Joined: Thu Feb 09, 2012 8:12 pm
Location: Cincinnati, OH

Re: The Efficient Frontier of Stocks and Bonds

Post by TheCluelessInvestor » Fri Apr 13, 2012 9:55 am

FinanceFun wrote:That answers my question. Theory is great, till you need to execute. Perhaps after a few years of execution, and accumulating an investment portfolio of some size, you will come to understand.


What's your alternative? Active managers underperform the market 2 out of 3 times.

FinanceFun
Posts: 722
Joined: Mon Nov 28, 2011 9:29 am

Re: The Efficient Frontier of Stocks and Bonds

Post by FinanceFun » Fri Apr 13, 2012 9:57 am

TheCluelessInvestor wrote:
FinanceFun wrote:That answers my question. Theory is great, till you need to execute. Perhaps after a few years of execution, and accumulating an investment portfolio of some size, you will come to understand.


What's your alternative? Active managers underperform the market 2 out of 3 times.


The alternative is a well thought out AA inclusive of stocks, bonds, REITs, and other non correlated assets

User avatar
Lbill
Posts: 4997
Joined: Thu Mar 13, 2008 11:25 pm
Location: Somewhere between Up and Down

Re: The Efficient Frontier of Stocks and Bonds

Post by Lbill » Fri Apr 13, 2012 10:03 am

Clueless - Are you able to share how you do those neat graphs and how you're able to display them in your posts? Really cool.
"Life can only be understood backward; but it must be lived forward." ~ Søren Kierkegaard | | "You can't connect the dots looking forward; but only by looking backwards." ~ Steve Jobs

TheCluelessInvestor
Posts: 46
Joined: Thu Feb 09, 2012 8:12 pm
Location: Cincinnati, OH

Re: The Efficient Frontier of Stocks and Bonds

Post by TheCluelessInvestor » Fri Apr 13, 2012 10:08 am

Lbill wrote:Clueless - Are you able to share how you do those neat graphs and how you're able to display them in your posts? Really cool.


I get the numbers out of an application at work (Morningstar EnCorr). Export them out to Excel and build the graph. Print Screen paste into Paint (I'm old schoold), Then upload it to http://www.fotoshack.us/.

No clue if that is the best way to do it. Learned it a couple days ago.

TheCluelessInvestor
Posts: 46
Joined: Thu Feb 09, 2012 8:12 pm
Location: Cincinnati, OH

Re: The Efficient Frontier of Stocks and Bonds

Post by TheCluelessInvestor » Fri Apr 13, 2012 10:09 am

FinanceFun wrote:The alternative is a well thought out AA inclusive of stocks, bonds, REITs, and other non correlated assets


What do you do when the markets tank? Add more bonds? What about when they rise? Add more equities?

rkhusky
Posts: 4545
Joined: Thu Aug 18, 2011 8:09 pm

Re: The Efficient Frontier of Stocks and Bonds

Post by rkhusky » Fri Apr 13, 2012 11:06 am

TheCluelessInvestor wrote:
rkhusky wrote:I am also suspicious of analysis that uses a single start or end date. Averaging over a range of end dates from 1992 - 2012 would make me more comfortable.


Six different time periods across two different end dates. I don't think of running an analysis that runs for 10, 20, or 30 years as picking specific time periods to make it look better.


I wonder how your results would have changed if the end point was 3/31/2009? or even 3/31/2010?

Even though it is common, I am also leery of using SD as a measure of risk without knowing the average duration of market drops.

ourbrooks
Posts: 1575
Joined: Fri Nov 13, 2009 4:56 pm

Re: The Efficient Frontier of Stocks and Bonds

Post by ourbrooks » Fri Apr 13, 2012 11:30 am

Am I correct that all of these analyses presume a single investment at the beginning of the time period and no withdrawals?

What happens during the decumulation phase, when you are making period withdrawals?

The safe withdrawal rate studies sumarized in the Wiki (http://www.bogleheads.org/wiki/Trinity_study_update) suggest that, to support a 4% inflation adjusted withdrawal rate, something between 40% and 60% stocks is optimal. A higher percentage of stocks actually increases the chances of running out of money as does a higher percentage of bonds. (Those who practice "age in bonds" need to reduce their withdrawal rate to 3% or 3.5% to maintain safety.)

Are there really two different answers to the optimum allocation question, one during accumulation and one during decumulation?

TheCluelessInvestor
Posts: 46
Joined: Thu Feb 09, 2012 8:12 pm
Location: Cincinnati, OH

Re: The Efficient Frontier of Stocks and Bonds

Post by TheCluelessInvestor » Fri Apr 13, 2012 11:35 am

rkhusky wrote:I wonder how your results would have changed if the end point was 3/31/2009? or even 3/31/2010?


Once again, as long as you are investing for the long term, you would've done just fine (returns):

Image

Image

Image

rkhusky wrote:Even though it is common, I am also leery of using SD as a measure of risk without knowing the average duration of market drops.


http://en.wikipedia.org/wiki/Efficient_frontier

TheCluelessInvestor
Posts: 46
Joined: Thu Feb 09, 2012 8:12 pm
Location: Cincinnati, OH

Re: The Efficient Frontier of Stocks and Bonds

Post by TheCluelessInvestor » Fri Apr 13, 2012 11:41 am

ourbrooks wrote:Am I correct that all of these analyses presume a single investment at the beginning of the time period and no withdrawals?

What happens during the decumulation phase, when you are making period withdrawals?

The safe withdrawal rate studies sumarized in the Wiki (http://www.bogleheads.org/wiki/Trinity_study_update) suggest that, to support a 4% inflation adjusted withdrawal rate, something between 40% and 60% stocks is optimal. A higher percentage of stocks actually increases the chances of running out of money as does a higher percentage of bonds. (Those who practice "age in bonds" need to reduce their withdrawal rate to 3% or 3.5% to maintain safety.)

Are there really two different answers to the optimum allocation question, one during accumulation and one during decumulation?


The idea here is not that this is a retirement fund. It is simply how beneficial bonds may or may not be during a long period of time.

ourbrooks
Posts: 1575
Joined: Fri Nov 13, 2009 4:56 pm

Re: The Efficient Frontier of Stocks and Bonds

Post by ourbrooks » Fri Apr 13, 2012 12:19 pm

TheCluelessInvestor wrote:
ourbrooks wrote:Am I correct that all of these analyses presume a single investment at the beginning of the time period and no withdrawals?

What happens during the decumulation phase, when you are making period withdrawals?

The safe withdrawal rate studies sumarized in the Wiki (http://www.bogleheads.org/wiki/Trinity_study_update) suggest that, to support a 4% inflation adjusted withdrawal rate, something between 40% and 60% stocks is optimal. A higher percentage of stocks actually increases the chances of running out of money as does a higher percentage of bonds. (Those who practice "age in bonds" need to reduce their withdrawal rate to 3% or 3.5% to maintain safety.)

Are there really two different answers to the optimum allocation question, one during accumulation and one during decumulation?


The idea here is not that this is a retirement fund. It is simply how beneficial bonds may or may not be during a long period of time.


It's very important that the long period of time be without withdrawals. My conclusion is that, before the start of retirement, a modest amount of bonds can reduce overall risk without too much of an impact on returns. After retirement starts, though, you need a much higher percentage of bonds to reduce risk.
Thirty years before retirement, something like 20% bonds works well; for the thirty years after retirement starts, raise that to 40% or 50% bonds if Wade Pfau's results are to be believied.

FinanceFun
Posts: 722
Joined: Mon Nov 28, 2011 9:29 am

Re: The Efficient Frontier of Stocks and Bonds

Post by FinanceFun » Fri Apr 13, 2012 12:21 pm

TheCluelessInvestor wrote:
FinanceFun wrote:The alternative is a well thought out AA inclusive of stocks, bonds, REITs, and other non correlated assets


What do you do when the markets tank? Add more bonds? What about when they rise? Add more equities?


Rebalance.

TheCluelessInvestor
Posts: 46
Joined: Thu Feb 09, 2012 8:12 pm
Location: Cincinnati, OH

Re: The Efficient Frontier of Stocks and Bonds

Post by TheCluelessInvestor » Fri Apr 13, 2012 12:38 pm

FinanceFun wrote:Rebalance.


Read my very first posting again.

FinanceFun
Posts: 722
Joined: Mon Nov 28, 2011 9:29 am

Re: The Efficient Frontier of Stocks and Bonds

Post by FinanceFun » Fri Apr 13, 2012 12:43 pm

TheCluelessInvestor wrote:
FinanceFun wrote:Rebalance.


Read my very first posting again.


What does that have to do with the question about how I handle the situation?

And out of curiosity... How do you plan on rebalancing a 100% stock portfolio?

TheCluelessInvestor
Posts: 46
Joined: Thu Feb 09, 2012 8:12 pm
Location: Cincinnati, OH

Re: The Efficient Frontier of Stocks and Bonds

Post by TheCluelessInvestor » Fri Apr 13, 2012 12:55 pm

FinanceFun wrote:And out of curiosity... How do you plan on rebalancing a 100% stock portfolio?


Small/large cap? Value/growth? Domestic/foreign? REIT?

TheCluelessInvestor
Posts: 46
Joined: Thu Feb 09, 2012 8:12 pm
Location: Cincinnati, OH

Re: The Efficient Frontier of Stocks and Bonds

Post by TheCluelessInvestor » Fri Apr 13, 2012 1:02 pm

ourbrooks wrote:
TheCluelessInvestor wrote:
ourbrooks wrote:Am I correct that all of these analyses presume a single investment at the beginning of the time period and no withdrawals?

What happens during the decumulation phase, when you are making period withdrawals?

The safe withdrawal rate studies sumarized in the Wiki (http://www.bogleheads.org/wiki/Trinity_study_update) suggest that, to support a 4% inflation adjusted withdrawal rate, something between 40% and 60% stocks is optimal. A higher percentage of stocks actually increases the chances of running out of money as does a higher percentage of bonds. (Those who practice "age in bonds" need to reduce their withdrawal rate to 3% or 3.5% to maintain safety.)

Are there really two different answers to the optimum allocation question, one during accumulation and one during decumulation?


The idea here is not that this is a retirement fund. It is simply how beneficial bonds may or may not be during a long period of time.


It's very important that the long period of time be without withdrawals. My conclusion is that, before the start of retirement, a modest amount of bonds can reduce overall risk without too much of an impact on returns. After retirement starts, though, you need a much higher percentage of bonds to reduce risk.
Thirty years before retirement, something like 20% bonds works well; for the thirty years after retirement starts, raise that to 40% or 50% bonds if Wade Pfau's results are to be believied.


Another sensible post. I’m not sure that 40% or 50% in bonds after retirement is the right number. It may be. But I guess it all depends on how much money you have and how much income you need. It may be that you need more or less bonds. There’s no simple answer. My idea was geared more towards someone who is younger and is saving for the longer term. There’s all kinds of recommended bond allocations out there. It may be that for a very long horizon, having bonds MAY be detrimental. There's no way to know for sure since nobody know what will happen with the market tomorrow, let alone 30 years from now.

FinanceFun
Posts: 722
Joined: Mon Nov 28, 2011 9:29 am

Re: The Efficient Frontier of Stocks and Bonds

Post by FinanceFun » Fri Apr 13, 2012 1:06 pm

TheCluelessInvestor wrote:
FinanceFun wrote:And out of curiosity... How do you plan on rebalancing a 100% stock portfolio?


Small/large cap? Value/growth? Domestic/foreign? REIT?


So you do sector and market cap bets? You don't believe in market weighting? How do you time your bets?

How about for clarity you POST your recommended Asset Allocation. Take the guesswork out?

Rodc
Posts: 13601
Joined: Tue Jun 26, 2007 9:46 am

Re: The Efficient Frontier of Stocks and Bonds

Post by Rodc » Fri Apr 13, 2012 1:32 pm

rkhusky wrote:Even though it is common, I am also leery of using SD as a measure of risk without knowing the average duration of market drops
.



http://en.wikipedia.org/wiki/Efficient_frontier


I think you will find that everyone posting already knows that.

Modern finance, for example the lifecycle approach that Bobk frequently posts about, as well as posts from his frequent opponent, Dick Purcell (names given to make it easier for you to search the site), and many more, explains why rkhusky is leery of using SD.

SD of monthly or annual returns was originally used because it makes the optimization codes linear, and hence just barely doable on 1950s computers (something more sophisticated was entire unworkable in the 1950s), not because it was a great measure of risk.

Now I would say for abstract work it is not horrible, but for individual planning especially late in accumulation and in retirement more is needed. Though given modern computing power it would seem at least modifying for the non-linear effects of losing and gaining money would be much better.
Last edited by Rodc on Fri Apr 13, 2012 2:29 pm, edited 1 time in total.
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.

555
Posts: 4955
Joined: Thu Dec 24, 2009 7:21 am

Re: The Efficient Frontier of Stocks and Bonds

Post by 555 » Fri Apr 13, 2012 2:13 pm

TheCluelessInvestor wrote:`I also wondered about the “your age in bonds” rule.'

TheCluelessInvestor wrote:`The idea here is not that this is a retirement fund. It is simply how beneficial bonds may or may not be during a long period of time.'

You need to be clear about what your topic is. There is a big difference between a retirement plan with accumulation and decumulation phases, and an eternal portfolio that is never even spent from.

Even for an endownment type fund (eternal but is spent from) they don't go anything like 100% stocks.

The pattern of contributions and withdrawals makes a huge difference to how you construct your portfolio.

User avatar
Random Musings
Posts: 5209
Joined: Thu Feb 22, 2007 4:24 pm
Location: Pennsylvania

Re: The Efficient Frontier of Stocks and Bonds

Post by Random Musings » Fri Apr 13, 2012 2:44 pm

For people who were in the market for a long time and had (most likely) the largest amount of their assets during the 2000-2010 timeframe, they felt the underperformance of equities. Timing can be a factor.

For people who were doing a stronger majority of saving in the 90's, well the market was good. Then, they could have taken some risk off the table - again, not bad timing as bonds still had a bull run the next decade. Again, timing can be a factor.

I don't think those 20 or 30 year curve charts mean much for the majority of investors since:

- For many people, the majority of savings are around a more concentrated time period (typically, closer to retirement)
- Hence, the 10 yr curves drive portfolio value more.

RM
I figure the odds be fifty-fifty I just might have something to say. FZ

User avatar
czeckers
Posts: 934
Joined: Thu May 17, 2007 3:49 pm
Location: Upstate NY

Re: The Efficient Frontier of Stocks and Bonds

Post by czeckers » Fri Apr 13, 2012 2:52 pm

@ CluelessInvestor:

I think your analysis is depends on the premise that history will repeat itself and that is simply not true. Your results are completely dependent on the time period chosen. Looking at the 10-year periods between your two graphs, you see that in the decade ending in 2012, 100% bonds did better than 100% stocks, but in the decade ending in 2002, the opposite is true. Using that information, how will you constuct your portfolio for the next decade?

Now if you are young, you may have 30 years until retirement and thus will have plenty of time to recover from any losses before retirement. However, what if you have a large loss 10 years before retirement? 5 years? 1 year? For a stock-heavy portfolio, that is REAL risk. In fact, I know many of my colleagues who were very stock heavy going into 2000 and were near retirement, and lost 1/2 their portfolio. At that point they had no choice but to postpone retirement and work an extra 3-5 years until their portfolios bounced back. That's real risk.

You have to understand that it can take a full decade to recover from a major market crash. If that's the case, you should seriously consider a much more conservative allocation at least 10 years out from retirement. If you decide to go 100% stocks right up until retirement, you must be prepared for the real possibility that a crash will hit in the last decade before your retirement, and you may end up having to postpone it for a while.

The other thing you are ignoring are withdrawals during retirement. There is a large difference in portfolio behavior between the accumulation phase and draw-down phase. If you have a market crash during the draw-down phase, particularly early in retirement, it will have a significant negative impact on the odds that you may in fact outlive your portfolio.

Finally, you can't underestimate the emotional impact of a market crash and the resultant effect on your portfolio. In the beginning, when the portfolio is relatively small, a 50% drop in the market may only represent the loss of ~1 year's worth of income in portfolio value, and, given the fact that you may have another 30 years until retirement, that loss doesn't feel as painful as there is plenty of time for either the markets to recover, or for you to make up the losses with additional savings. However, as you near retirement, that same crash could well represent a decade's worth of income. However, with advanced age, perhaps uncertain health, and a limited time horizon, such a crash would have a very large emotional impact.

Food for thought.

-K
The Espresso portfolio: | | 16% LCV, 16% SCV, 16% EM, 8% Int'l Value, 8% Int'l Sm, 8% US REIT, 8% Int'l REIT, 20% Inter-term US Treas | | "A journey of a thousand miles begins with a single step."

Post Reply