Making a case for bonds (All About Asset Allocation)

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
Post Reply
Topic Author
aaronh
Posts: 17
Joined: Sat Feb 22, 2014 8:17 am

Making a case for bonds (All About Asset Allocation)

Post by aaronh »

Hello!

I am 38 and have been investing via work sponsored 401K plans for the last decade. I've made some terrible mistakes along the way, and figured it was time to educate myself and take more control of my financial future.

I began my education by simply speaking with friends/co-workers who I considered more knowledgable about investing than myself. Through various discussions, I came to the following conclusion:

1.) My wife and I have very stable careers.
2.) We are maxing out our work sponsored 401K plans and ROTH IRAs.
3.) We are contributing to our childrens' edvest accounts.
4.) We have 6 months of expenses saved.
5.) We have 20+ years until retirement.
6.) We pray for significant drops in stocks so that we can acquire them up at low prices.

leading to...

7.) We have nothing but time on our sides, so we should be investing in 100% equities.

Well, so I thought, until I read All About Asset Allocation, by Richard Ferri. In his book, he makes a compelling case for even the most aggressive investors to contain at least 20% bonds in their portfolio. I'm hoping that someone could provide clarification on his arguments? Below are his thoughts on making his case.

First argument
when stocks fall in value, investors should take that opportunity to buy more stocks. a 100 percent stock portfolio precludes this from happening. A 20 percent bond allocation will allow stocks to be purchased in a down market.
This is an intriguing statement, but I can't help but feel as though he's suggesting that I should be market timing somehow. Or is he suggesting that simple rebalancing will achieve this effect of buying low and selling high? It’s not very clear.

Second argument
higher volatility of returns leads to lower compounded returns and vice versa. Accordingly, any strategy that lowers the return volatility of the portfolio without lowering the simple average return will increase the compounded return.
If I understand correctly, he's suggesting that negatively correlated assets help reduce risk, while increasing returns. Or in other words, by dampening the market highs and lows, you stand to gain more via compounded returns over the long run. This part confuses me. Could someone shed light as to how this compounding of returns works by compressing the highs and lows?

I'd really like to understand more about his arguments before ditching my 100% equity allocation strategy and adopting an 80/20 stock/bond allocation strategy.

Thanks so much for taking the time to read this post and potentially offer any useful insight!
hoops777
Posts: 4603
Joined: Sun Apr 10, 2011 12:23 pm
Location: Behind the 3 point line

Re: Making a case for bonds (All About Asset Allocation)

Post by hoops777 »

This topic has been covered many times.Search around on this site and you will find all of your answers.
K.I.S.S........so easy to say so difficult to do.
staythecourse
Posts: 6993
Joined: Mon Jan 03, 2011 8:40 am

Re: Making a case for bonds (All About Asset Allocation)

Post by staythecourse »

Welcome aboard.

To your first question: Yes rebalancing is market timing, but good market timing if one believes (which most of us do) that price movements in asset classes, like stocks and bonds, have reversion to mean, i.e. what goes up comes back down and vice versa. So it is an easy way to capture the sell high and buy low concept. Now the more you read you will learn the idea of rebalancing has NOTHING to do with that, but allows an investor a mechanism to maintain as steady volatility he chooses. For example: If stocks are booming naturally you portfolio asset allocation gets tilted to more stocks and thus the portfolio becomes more volatile without you intentionally doing so. Ask the near retirees how that worked out in late 90's and prior to the plunge in 2008.

Now the quote from Mr. Ferri ONLY makes sense If one accepts the theory that an 80/20 portfolio rebalanced when markets go down the toilet (thus allowing to buy stocks cheaper) will OUTPEFORM a 100% equity portfolio continually invested rain or shine? To my limited knowledge this simple question has not been researched oddly enough using historical data. I believe it was Nisi who made a very good observation once that all rebalancing data is based on Jan. 1st mechanically without thought of how the market is doing thus NOT trying to exploit the "sell high and buy low" theory.

To question 2: This is, in my opinion, the largest single argument in modern portfolio theory that resides among us amateur investing geeks. To rephrase the question does adding asset classes that have decreased expected returns decrease volatility and thus increase compounded returns return BETTER then having higher returning and volatile asset classes alone. I am part way through a book that may have just answered that question in regards to stocks and bonds.

The answer seems to be adding a lower returning lower volatility asset class (bonds) will ONLY improve the returns IF the rebalancing bonus by adding into the portfolio is GREATER then the equity risk premium. So if adding bonds to stocks and the rebalancing bonus when all is done, for example, 0.5% and the ERP ended up being 2% then the answer is no. So most of the time it plays out how it should: Adding bonds decreases volatility of the portfolio, but also lowers return. The only way to not seems to be: 1. If one adds the holy grail of investing: adding an asset class with equity like returns, high volatility, and with zero to neg. correlation or 2. a period like 2000's occurs where the rebalancing bonus was far greater then the ERP (which I am guessing was negative actually).

Hope this helps.

Good luck.
"The stock market [fluctuation], therefore, is noise. A giant distraction from the business of investing.” | -Jack Bogle
User avatar
packer16
Posts: 1488
Joined: Sat Jan 04, 2014 1:28 pm

Re: Making a case for bonds (All About Asset Allocation)

Post by packer16 »

I am about 10 years older than you and have been 100% equity since I have started investing in the 1990s. I think real question is can you deal with the volatility of stocks? Since you have been through 2008, I assume you stayed the course. If you goal is to maximize returns, you have a 20 year time horizon, you appear to be dollar cost averaging and you will not panic in declines then I think 100% equity is fine. As you approach the time where your investments have to provide income then bonds will become more important. The allocation to bonds for some provides comfort and will allow them to stay the course, for those folks bonds make sense as the volatility may cause a panic sale (the worst situation to be in).

The issue with the allocation to bonds is the opportunity cost of investing in equity which on average is the ERP or about 4 to 5% annually. This is not much for year but will add up if you plan on holding assets for 20+ years.

Packer
Buy cheap and something good might happen
Topic Author
aaronh
Posts: 17
Joined: Sat Feb 22, 2014 8:17 am

Re: Making a case for bonds (All About Asset Allocation)

Post by aaronh »

staythecourse, thanks for the warm welcome and well thought out answers!
To your first question: Yes rebalancing is market timing, but good market timing if one believes (which most of us do) that price movements in asset classes, like stocks and bonds, have reversion to mean, i.e. what goes up comes back down and vice versa. So it is an easy way to capture the sell high and buy low concept. Now the more you read you will learn the idea of rebalancing has NOTHING to do with that, but allows an investor a mechanism to maintain as steady volatility he chooses. For example: If stocks are booming naturally you portfolio asset allocation gets tilted to more stocks and thus the portfolio becomes more volatile without you intentionally doing so. Ask the near retirees how that worked out in late 90's and prior to the plunge in 2008.

Now the quote from Mr. Ferri ONLY makes sense If one accepts the theory that an 80/20 portfolio rebalanced when markets go down the toilet (thus allowing to buy stocks cheaper) will OUTPEFORM a 100% equity portfolio continually invested rain or shine? To my limited knowledge this simple question has not been researched oddly enough using historical data. I believe it was Nisi who made a very good observation once that all rebalancing data is based on Jan. 1st mechanically without thought of how the market is doing thus NOT trying to exploit the "sell high and buy low" theory.
I think I see what you're saying. What I find most confusing about all of this is that throughout the book, he states (repeatedly) that there is no such thing as perfectly negatively correlated asset classes. So making an assumption that bonds will in some way allow me to buy low and sell high (due to negative correlations) may really never happen (or at least as frequently as I would hope).

In addition, there are times when negative correlations exist between equity classes (large, mid, small cap, etc.). For example, in The Four Pillars of Investing, by William Bernstein, there is a chart that states in 1998, U.S. Large Stocks (S&P 500) were up 28.58%, while U.S. Small Stocks (CRSP 9-10) were down -7.3 %. So it seems to me that negative correlations are possible between all asset classes. I guess this is part of my confusion. Perhaps historically, bonds have been more negatively correlated with stocks than differing equity types? But if I'm not mistaken, it seems that even having 100% equities can afford me the luxury to buy low and sell high, yes? For example, in 98', reallocating would have allowed me to purchase small cap stocks at a discount rate and sell large cap stocks at an inflated rate.
To question 2: This is, in my opinion, the largest single argument in modern portfolio theory that resides among us amateur investing geeks. To rephrase the question does adding asset classes that have decreased expected returns decrease volatility and thus increase compounded returns return BETTER then having higher returning and volatile asset classes alone. I am part way through a book that may have just answered that question in regards to stocks and bonds.

The answer seems to be adding a lower returning lower volatility asset class (bonds) will ONLY improve the returns IF the rebalancing bonus by adding into the portfolio is GREATER then the equity risk premium. So if adding bonds to stocks and the rebalancing bonus when all is done, for example, 0.5% and the ERP ended up being 2% then the answer is no. So most of the time it plays out how it should: Adding bonds decreases volatility of the portfolio, but also lowers return. The only way to not seems to be: 1. If one adds the holy grail of investing: adding an asset class with equity like returns, high volatility, and with zero to neg. correlation or 2. a period like 2000's occurs where the rebalancing bonus was far greater then the ERP (which I am guessing was negative actually).
You've rephrased my question beautifully, which lets me know that you truly understand what I'm after. Embarrassingly, I've read your statement a few times and I'm still confused. I'm still not sure how the compounding aspect works. It's hard for me to wrap my head around how making less (due to picking less volatile asset classes) can make you more in the long run due to compounding.

Thanks again for your time and thoughts!
Topic Author
aaronh
Posts: 17
Joined: Sat Feb 22, 2014 8:17 am

Re: Making a case for bonds (All About Asset Allocation)

Post by aaronh »

packer16 wrote:I am about 10 years older than you and have been 100% equity since I have started investing in the 1990s. I think real question is can you deal with the volatility of stocks? Since you have been through 2008, I assume you stayed the course. If you goal is to maximize returns, you have a 20 year time horizon, you appear to be dollar cost averaging and you will not panic in declines then I think 100% equity is fine. As you approach the time where your investments have to provide income then bonds will become more important. The allocation to bonds for some provides comfort and will allow them to stay the course, for those folks bonds make sense as the volatility may cause a panic sale (the worst situation to be in).

The issue with the allocation to bonds is the opportunity cost of investing in equity which on average is the ERP or about 4 to 5% annually. This is not much for year but will add up if you plan on holding assets for 20+ years.
Hi packer16, thanks for the response!

I do indeed plan on plan on staying the course, and in fact, stayed the course in 2008. I don't see this changing. In fact, more recently (based on educating myself from books), I have been greedily hoping for more opportunities to buy stocks on the cheap.

I guess what I'm most concerned about is not having my head around the concept of having less volatile asset classes to help suppress the market highs and lows, eventually leading to higher compounding returns. If doing so truly gives me higher returns over time, it seems like having bonds in my portfolio is a no-brainer.

However, I can't adopt a strategy if I don't fully understand it, and I'm still not seeing how it's possible to make more with less (having bonds in my portfolio). Does this make sense? I feel like I'm having a hard time relaying my thoughts unfortunately. Perhaps I need to give the book another read...

Thanks again for your thoughts!
User avatar
dodecahedron
Posts: 6607
Joined: Tue Nov 12, 2013 11:28 am

Re: Making a case for bonds (All About Asset Allocation)

Post by dodecahedron »

aaronh wrote:Embarrassingly, I've read your statement a few times and I'm still confused. I'm still not sure how the compounding aspect works. It's hard for me to wrap my head around how making less (due to picking less volatile asset classes) can make you more in the long run due to compounding.

Thanks again for your time and thoughts!
A simple (if unrealistic) numerical example may help:

portfolio A alternates between losing 10% and gaining 25% in even and odd years, for an average return of 7.5%.

portfolio B consistently gains 7% per year.

Start with $100K in each portfolio.

After one year, portfolio A is worth 0.9*100K= $90K. After two years, portfolio A is worth 1.25*90K=$112.5K.

After one year, portfolio B is worth 1.07*100K=$107K. After two years, portfolio B is worth 1.07*107K=$114.49K.

So after a cycle of two years, portfolio B is ahead by almost $2,000, even though its (arithmetic) average return is less than portfolio A's average. And this difference will continue to compound over time and the divergence will increase.
zerlegen
Posts: 56
Joined: Thu Jun 04, 2009 9:57 am

Re: Making a case for bonds (All About Asset Allocation)

Post by zerlegen »

I like this example. So intuitively, by dampening the volatility, you're actually amplifying the effect of compound interest.
staythecourse
Posts: 6993
Joined: Mon Jan 03, 2011 8:40 am

Re: Making a case for bonds (All About Asset Allocation)

Post by staythecourse »

zerlegen wrote:I like this example. So intuitively, by dampening the volatility, you're actually amplifying the effect of compound interest.
Perfect way to articulate it. The math geek stuff is by decreasing the volatility of the portfolio one approximates the CAGR (geometric mean) closer to the average annual return (arithmetic mean). This the whole goal of modern portfolio theory.

Good luck.
"The stock market [fluctuation], therefore, is noise. A giant distraction from the business of investing.” | -Jack Bogle
Topic Author
aaronh
Posts: 17
Joined: Sat Feb 22, 2014 8:17 am

Re: Making a case for bonds (All About Asset Allocation)

Post by aaronh »

dodecahedron wrote:
aaronh wrote:Embarrassingly, I've read your statement a few times and I'm still confused. I'm still not sure how the compounding aspect works. It's hard for me to wrap my head around how making less (due to picking less volatile asset classes) can make you more in the long run due to compounding.

Thanks again for your time and thoughts!
A simple (if unrealistic) numerical example may help:

portfolio A alternates between losing 10% and gaining 25% in even and odd years, for an average return of 7.5%.

portfolio B consistently gains 7% per year.

Start with $100K in each portfolio.

After one year, portfolio A is worth 0.9*100K= $90K. After two years, portfolio A is worth 1.25*90K=$112.5K.

After one year, portfolio B is worth 1.07*100K=$107K. After two years, portfolio B is worth 1.07*107K=$114.49K.

So after a cycle of two years, portfolio B is ahead by almost $2,000, even though its (arithmetic) average return is less than portfolio A's average. And this difference will continue to compound over time and the divergence will increase.
Brilliant, this makes perfect sense! It's interesting that out of the 4 books that I've recently read (All About Asset Allocation, The Four Pillars of Investing, Asset Allocation - Balancing Financial Risk, and The Boggleheads' Guide to Investing (which directed me to this site :happy ), only one of them discussed the benefit of compounding returns with regard to reducing portfolio volatility with bonds. Admittedly, I've read through all 4 pretty quickly and I may have breezed over this concept in one of the others.
User avatar
kenyan
Posts: 3015
Joined: Wed Jan 12, 2011 11:16 pm

Re: Making a case for bonds (All About Asset Allocation)

Post by kenyan »

aaronh wrote:
In addition, there are times when negative correlations exist between equity classes (large, mid, small cap, etc.). For example, in The Four Pillars of Investing, by William Bernstein, there is a chart that states in 1998, U.S. Large Stocks (S&P 500) were up 28.58%, while U.S. Small Stocks (CRSP 9-10) were down -7.3 %. So it seems to me that negative correlations are possible between all asset classes. I guess this is part of my confusion. Perhaps historically, bonds have been more negatively correlated with stocks than differing equity types? But if I'm not mistaken, it seems that even having 100% equities can afford me the luxury to buy low and sell high, yes? For example, in 98', reallocating would have allowed me to purchase small cap stocks at a discount rate and sell large cap stocks at an inflated rate.
Don't count on this too often. Correlations across all types of equities are much, much higher than correlations between bonds and equities (>0.8 vs near-zero or negative). Recent experience also taught us that in a true financial crisis, correlations among risky asset classes basically go to 1, while bonds were off doing their own thing. Every equity type - small, large, growth, value, international, domestic, REIT - dropped precipitously in 2008-early 2009; high-quality bonds rose.

If you want low correlations but equity-like returns, consider adding REITs. If you want negative correlations, add long-term treasuries. High-quality intermediates will probably give you correlations near zero.
Retirement investing is a marathon.
Costanza
Posts: 52
Joined: Sun Dec 08, 2013 7:08 pm

Re: Making a case for bonds (All About Asset Allocation)

Post by Costanza »

Most renderings of the "efficient frontier" generated by modern portfolio theory -- at least the several that I have seen in books and research -- indicate that you do not increase returns by adding bonds. When a portfolio moves from 100% to 80% equities, expected return falls a little (and expected volatility falls a good deal). This should be desirable to you only if you value the (larger) decrease in volatility more than the (smaller) decrease in return.

The free lunch of diversification occurs only when you add stocks to bonds. By having, let's say, 30% equities and 70% fixed income rather than 100% fixed income, you increase expected return without increasing volatility. But, to my knowledge, most experts don't believe you can increase return by adding bonds to any portfolio, unless those bonds help you stay the course during a stock market plunge.

(By the way, further free lunches can occur from adding additional asset classes like international stocks and REITs. But I absolutely wouldn't think of U.S. small caps and mid caps as diversifiers; they usually correlate very positively with U.S. large caps, notwithstanding some exceptional years.)
Topic Author
aaronh
Posts: 17
Joined: Sat Feb 22, 2014 8:17 am

Re: Making a case for bonds (All About Asset Allocation)

Post by aaronh »

Costanza wrote:Most renderings of the "efficient frontier" generated by modern portfolio theory -- at least the several that I have seen in books and research -- indicate that you do not increase returns by adding bonds. When a portfolio moves from 100% to 80% equities, expected return falls a little (and expected volatility falls a good deal). This should be desirable to you only if you value the (larger) decrease in volatility more than the (smaller) decrease in return.
See, that is what I thought too, but the "compounding" keyword changes my view on that now. I think it comes down to simple vs. compounded returns, the latter being what we really care about in the long run, no? Based on this answer, I'm becoming convinced that adding bonds to a portfolio can indeed improve your overall returns in the long run. However, seeing as only one of 4 investing books that I've read suggest adding bonds to improve returns, I'm not so certain.

Signed, still confused...
Costanza
Posts: 52
Joined: Sun Dec 08, 2013 7:08 pm

Re: Making a case for bonds (All About Asset Allocation)

Post by Costanza »

aaronh wrote:
Costanza wrote:Most renderings of the "efficient frontier" generated by modern portfolio theory -- at least the several that I have seen in books and research -- indicate that you do not increase returns by adding bonds. When a portfolio moves from 100% to 80% equities, expected return falls a little (and expected volatility falls a good deal). This should be desirable to you only if you value the (larger) decrease in volatility more than the (smaller) decrease in return.
See, that is what I thought too, but the "compounding" keyword changes my view on that now. I think it comes down to simple vs. compounded returns, the latter being what we really care about in the long run, no? Based on this answer, I'm becoming convinced that adding bonds to a portfolio can indeed improve your overall returns in the long run. However, seeing as only one of 4 investing books that I've read suggest adding bonds to improve returns, I'm not so certain.

Signed, still confused...
I suspect the way to reconcile these considerations lies in the flaws of this post. Portfolio A outperforms portfolio B, true. But by adding 20% bonds to an all-stock portfolio, you're not going to eliminate volatility completely, as the post assumes. You will, rather, limit it slightly -- and, according to most research, not enough to increase return. Bonds have volatility themselves, and making them 20% of your portfolio is unlikely to have a big impact on either returns or volatility.
User avatar
Wildebeest
Posts: 1204
Joined: Fri Dec 27, 2013 1:36 pm

Re: Making a case for bonds (All About Asset Allocation)

Post by Wildebeest »

100 % Stocks in US has had a higher yield than 80 %Stock 20 % Bond. However that is looking back. However nobody knows what the future will bring. If you are prudent you hedge your bets and diversify ( this includes bonds). In part because of the "inner demons" as outlined in The enemy in the mirror as per William Bernstein.

Just my two cents.
The Golden Rule: One should treat others as one would like others to treat oneself.
User avatar
packer16
Posts: 1488
Joined: Sat Jan 04, 2014 1:28 pm

Re: Making a case for bonds (All About Asset Allocation)

Post by packer16 »

The best way to increase returns if you are fine with 100% volatility is to tilt to small cap value with a SCV index fund and real estate. I think if you do tilt you need to ensure you are getting the tilt you expect. For example, currently the Vanguard SCV index fund has a higher PE than the Vanguard 500 fund. You can get low cost exposure via Vanguard funds.

Packer
Buy cheap and something good might happen
Topic Author
aaronh
Posts: 17
Joined: Sat Feb 22, 2014 8:17 am

Re: Making a case for bonds (All About Asset Allocation)

Post by aaronh »

packer16 wrote:The best way to increase returns if you are fine with 100% volatility is to tilt to small cap value with a SCV index fund and real estate. I think if you do tilt you need to ensure you are getting the tilt you expect. For example, currently the Vanguard SCV index fund has a higher PE than the Vanguard 500 fund. You can get low cost exposure via Vanguard funds.

Packer
That's good advice, I'll chew on that.
dbr
Posts: 46181
Joined: Sun Mar 04, 2007 8:50 am

Re: Making a case for bonds (All About Asset Allocation)

Post by dbr »

aaronh wrote:
packer16 wrote:The best way to increase returns if you are fine with 100% volatility is to tilt to small cap value with a SCV index fund and real estate. I think if you do tilt you need to ensure you are getting the tilt you expect. For example, currently the Vanguard SCV index fund has a higher PE than the Vanguard 500 fund. You can get low cost exposure via Vanguard funds.

Packer
That's good advice, I'll chew on that.
That isn't advice; it is a statement about the expected returns of different asset classes. What one does or should do with that information is an entirely different issue.
cb474
Posts: 905
Joined: Tue Jan 19, 2010 5:32 am

Re: Making a case for bonds (All About Asset Allocation)

Post by cb474 »

dbr wrote:
aaronh wrote:
packer16 wrote:The best way to increase returns if you are fine with 100% volatility is to tilt to small cap value with a SCV index fund and real estate. I think if you do tilt you need to ensure you are getting the tilt you expect. For example, currently the Vanguard SCV index fund has a higher PE than the Vanguard 500 fund. You can get low cost exposure via Vanguard funds.

Packer
That's good advice, I'll chew on that.
That isn't advice; it is a statement about the expected returns of different asset classes. What one does or should do with that information is an entirely different issue.
I've wondered, are PE ratios between different types of funds, in this case a large blend fund (based on the S&P 500) and a small value fund, directly comparable in a one-to-one manner like this? Do small value funds have the same historical mean for their PE as the S&P? Is this the relevant way to compare the expected returns of these different types of funds?
dbr
Posts: 46181
Joined: Sun Mar 04, 2007 8:50 am

Re: Making a case for bonds (All About Asset Allocation)

Post by dbr »

cb474 wrote:
That isn't advice; it is a statement about the expected returns of different asset classes. What one does or should do with that information is an entirely different issue.
I've wondered, are PE ratios between different types of funds, in this case a large blend fund (based on the S&P 500) and a small value fund, directly comparable in a one-to-one manner like this? Do small value funds have the same historical mean for their PE as the S&P? Is this the relevant way to compare the expected returns of these different types of funds?[/quote]

The definitive theory on the subject would be the work of Fama and French with the three factor model for investment returns (q.v.*). That research shows that returns of an asset class can be predicted to statistical accuracy that is high for financial work by regression on market risk, on a value factor (value stocks return more), and on a size factor (small stocks return more). The value metric in that theory is book to market ratio rather than P/E. It is a question of exploring around to find the greatest explanatory power. The general idea of the theory is to assume that return is related to risk although there is not a clearly understood rationale for why size and value are risks. In the theory it seems that one could say the definition of risk is "that which predicts return."

*q.v. = quod vide or, in English, which see, meaning to look up the subject in some appropriate place. I think there is a section on it in the Wiki.

I believe I am excused from defining PE and S&P, nor will I call the OP to account for using these terms, nor will I define OP or Wiki.

I hereby promise that this is the last post where I will beat this horse.
User avatar
packer16
Posts: 1488
Joined: Sat Jan 04, 2014 1:28 pm

Re: Making a case for bonds (All About Asset Allocation)

Post by packer16 »

Why wouldn't they be comparable? They are telling you how much you are paying for a dollar of earnings. I would want to pay less for smaller firms all things being equal. I think most of the time the SCV does have a lower P/E than the LCV as a premium is being paid for stability and liquidity. Given the parity, I would probably have less tilt than I would if SCV has a P/E less than LCV.

Packer
Buy cheap and something good might happen
Dandy
Posts: 6701
Joined: Sun Apr 25, 2010 7:42 pm

Re: Making a case for bonds (All About Asset Allocation)

Post by Dandy »

We have nothing but time on our sides, so we should be investing in 100% equities.
At 38 you have a lot of time but not that would automatically support all equities. If you use age in bonds as a beginning point then you are closer to the classic 60/40 allocation.

I'm not a fan of 100% equities. Maybe someone in their 20's or someone who has 5or 10 million but otherwise, for most of us, a more moderate allocation seems best. My guess is that people tend to overestimate their risk tolerance.

Fixed income provides some stability and the ability to buy equities when they are down. If you have a 70/30 portfolio and re balance based on some objective criteria you shouldn't think of it as market timing. You are just bringing your risk back to your original goal. You are not predicting the future. I would urge most investors to have a decent allocation to fixed income. That could be bonds/funds or stable value funds/CDs etc.
User avatar
abuss368
Posts: 27850
Joined: Mon Aug 03, 2009 2:33 pm
Location: Where the water is warm, the drinks are cold, and I don't know the names of the players!
Contact:

Re: Making a case for bonds (All About Asset Allocation)

Post by abuss368 »

Bonds provide excellent "dry powder" as Jack Bogle often notes. When the market tanks, and it always does, the flexibility of having bonds and being able to rebalance into equities (combined with new contributions) was invaluable. The period 2007 - 2009 really enforced this for us. In hindsight we are glad we had bonds and the opportunity to rebalance.
John C. Bogle: “Simplicity is the master key to financial success."
Topic Author
aaronh
Posts: 17
Joined: Sat Feb 22, 2014 8:17 am

Re: Making a case for bonds (All About Asset Allocation)

Post by aaronh »

Dandy wrote:Fixed income provides some stability and the ability to buy equities when they are down. If you have a 70/30 portfolio and re balance based on some objective criteria you shouldn't think of it as market timing. You are just bringing your risk back to your original goal. You are not predicting the future.
abuss368 wrote: Bonds provide excellent "dry powder" as Jack Bogle often notes. When the market tanks, and it always does, the flexibility of having bonds and being able to rebalance into equities (combined with new contributions) was invaluable. The period 2007 - 2009 really enforced this for us. In hindsight we are glad we had bonds and the opportunity to rebalance.
Thanks for the input. My head literally swirls from confusion. The entire stock/bond discussions/opinions seems to sway heavily in both directions. Both of your logic seems sound. But is this is making an assumption that there is always a negative correlation between stocks and bonds?

Dandy, According to R. Ferri in his Asset Allocation book, constant negatively correlated assets do not exist. So while you say "You are not predicting the future", doesn't it seem like we are predicting the future if we assume that stocks and bonds will have a negative correlation in the next 20+ years (my timeline)? I really want to understand, I feel as if there's a critical piece of information missing. Are we just saying that more times than not, there is a negative correlation between stocks and bonds?

What makes matters worse for me is that I've read this this thread, and just when I think I should be investing in bonds at some capacity, much of the content of the post causes me second guess myself.
User avatar
abuss368
Posts: 27850
Joined: Mon Aug 03, 2009 2:33 pm
Location: Where the water is warm, the drinks are cold, and I don't know the names of the players!
Contact:

Re: Making a case for bonds (All About Asset Allocation)

Post by abuss368 »

Benjamin Graham, the mentor of Warren Buffett, noted that no portfolio should have less than 25% in bonds!
John C. Bogle: “Simplicity is the master key to financial success."
Topic Author
aaronh
Posts: 17
Joined: Sat Feb 22, 2014 8:17 am

Re: Making a case for bonds (All About Asset Allocation)

Post by aaronh »

packer16 wrote:I am about 10 years older than you and have been 100% equity since I have started investing in the 1990s. I think real question is can you deal with the volatility of stocks? Since you have been through 2008, I assume you stayed the course. If you goal is to maximize returns, you have a 20 year time horizon, you appear to be dollar cost averaging and you will not panic in declines then I think 100% equity is fine. As you approach the time where your investments have to provide income then bonds will become more important. The allocation to bonds for some provides comfort and will allow them to stay the course, for those folks bonds make sense as the volatility may cause a panic sale (the worst situation to be in).

The issue with the allocation to bonds is the opportunity cost of investing in equity which on average is the ERP or about 4 to 5% annually. This is not much for year but will add up if you plan on holding assets for 20+ years.

Packer
Packer, I understand what you mean by the opportunity cost of investing in bonds. Being an investor with an all-equity portfolio, I would really appreciate your thoughts on the following quotes:
Dandy wrote: Fixed income provides some stability and the ability to buy equities when they are down. If you have a 70/30 portfolio and re balance based on some objective criteria you shouldn't think of it as market timing. You are just bringing your risk back to your original goal. You are not predicting the future.
abuss368 wrote: Bonds provide excellent "dry powder" as Jack Bogle often notes. When the market tanks, and it always does, the flexibility of having bonds and being able to rebalance into equities (combined with new contributions) was invaluable. The period 2007 - 2009 really enforced this for us. In hindsight we are glad we had bonds and the opportunity to rebalance.
It's hard for me to find balance between owning and not owning bonds. On the one hand, owning them could allow for me to buy stocks on the cheap by re-balancing during a down market. This assumes that there is a negative correlation between stocks and bonds. On the other hand, it's hard (impossible?) to be certain that a negative correlation between stocks and bonds will exist again in during my time frame, in which case, I potentially lost out on gains from stocks due to my bond ownership.

Sorry to beat a dead horse here, I really want to understand.
Last edited by aaronh on Mon Feb 24, 2014 12:15 pm, edited 1 time in total.
Rodc
Posts: 13601
Joined: Tue Jun 26, 2007 9:46 am

Re: Making a case for bonds (All About Asset Allocation)

Post by Rodc »

To your first question: Yes rebalancing is market timing, but good market timing if one believes (which most of us do) that price movements in asset classes, like stocks and bonds, have reversion to mean, i.e. what goes up comes back down and vice versa.
Lots of very long threads show both statements are not universally held to be true. Indeed rebalancing as market timing seems to be a minority view. I won't rehash the arguments as there are already many hundreds of posts on the topic.

We have a current thread on reversion to the mean that is worth checking into as well. Very unclear that RTM is true, though it may depend on what one means by the term as there are at least a few definitions so one needs an actual precise definition.

The locked thread on why own bonds at all (http://www.bogleheads.org/forum/viewtop ... 0&t=132539) has some good arguments for and against holding bonds and there is no need to read past the first page or two to see them. This might be of some value to the OP.
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.
placeholder
Posts: 8421
Joined: Tue Aug 06, 2013 12:43 pm

Re: Making a case for bonds (All About Asset Allocation)

Post by placeholder »

Historically about the most consistent (at least in direction if not magnitude) negative correlator for stocks has been long bonds which you can run some Morningstar charts to see.
User avatar
abuss368
Posts: 27850
Joined: Mon Aug 03, 2009 2:33 pm
Location: Where the water is warm, the drinks are cold, and I don't know the names of the players!
Contact:

Re: Making a case for bonds (All About Asset Allocation)

Post by abuss368 »

If the market decline and turned into a severe bear market for many years, could you stomach to see your equities cut in half or more?
John C. Bogle: “Simplicity is the master key to financial success."
Rodc
Posts: 13601
Joined: Tue Jun 26, 2007 9:46 am

Re: Making a case for bonds (All About Asset Allocation)

Post by Rodc »

FWIW: here is a look at US stocks and US treasuries over a fairly long time. It is a little out of date. But it shows that historically at least keeping some dry powder was not a great idea (but not bad either). Adding a little bit of stocks to a bond heavy portfolio was a great idea in that it really did improve risk adjusted return. Adding some bonds to a stock heavy portfolio simply reduced return commensurate with the reduction in risk (not at all bad, but nothing special).

FWIW: the lines are for various selected portfolios like equal parts TSM/Large value/Small value (3x33). The Blue region is all fixed portfolios.

Personally I don't trust that history will repeat (ie more stock risk than this may show up), and I'm getting old enough that "in the long term" may well be more long term than I have to ride out. So I have a solid amount of bonds. YMMV.


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

Re: Making a case for bonds (All About Asset Allocation)

Post by Rodc »

placeholder wrote:Historically about the most consistent (at least in direction if not magnitude) negative correlator for stocks has been long bonds which you can run some Morningstar charts to see.
But at the portfolio level you are better off intermediate.

Compare this graph to the previous one.

Image

A the high stock end (high risk end) the returns are about the same as using 10-year bonds. At the low end of bonds (low risk end) you get more risk reduction bang for your buck using safer bonds. Or so it seems.

That said, I'm not sure a reduction of a couple of percent in STD is that earth-shatteringly better (say look at the risk at a return of 7%).
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.
User avatar
hornet96
Posts: 811
Joined: Sun Nov 25, 2012 5:45 pm

Re: Making a case for bonds (All About Asset Allocation)

Post by hornet96 »

dodecahedron wrote:
aaronh wrote:Embarrassingly, I've read your statement a few times and I'm still confused. I'm still not sure how the compounding aspect works. It's hard for me to wrap my head around how making less (due to picking less volatile asset classes) can make you more in the long run due to compounding.

Thanks again for your time and thoughts!
A simple (if unrealistic) numerical example may help:

portfolio A alternates between losing 10% and gaining 25% in even and odd years, for an average return of 7.5%.

portfolio B consistently gains 7% per year.

Start with $100K in each portfolio.

After one year, portfolio A is worth 0.9*100K= $90K. After two years, portfolio A is worth 1.25*90K=$112.5K.

After one year, portfolio B is worth 1.07*100K=$107K. After two years, portfolio B is worth 1.07*107K=$114.49K.

So after a cycle of two years, portfolio B is ahead by almost $2,000, even though its (arithmetic) average return is less than portfolio A's average. And this difference will continue to compound over time and the divergence will increase.
To further illustrate the example above, the reason this happens is due to the difference between the geometric mean vs. the arithmetic mean. The geometric mean return for each of these examples is computed as follows:

Portfolio A: ((0.9 x 1.25)^(1/2)) - 1 = 6.01%

Portfolio B: ((1.07 x 1.07)^(1/2)) - 1 = 7.00%

Thus, while Portfolio A's arithmetic mean of 7.5% is greater than Portfolio B's of 7%, it's geometric mean of 6.01% is less than Portfolio B's of 7%.

Basically, an arithmetic mean assumes that all observations are independent of one another, while the geometric mean accounts for situations where the observations (i.e. portfolio returns & ending value) are not necessarily independent of each other (e.g. sequence of returns is important).

I believe this is basically the effect you are looking for as to how it could be possible to maximize expected return by adding a lower-returning asset class (bonds) with a corresponding reduction in portfolio volatility (compared to a 100% stock allocation). Obviously the example above is extreme, but illustrates the point that while Portfolio A was assumed to have a higher expected return at time zero, the zero-volatility Portfolio B left you with more money at the end of 2 years. As another poster mentioned earlier, this type of reasoning (along with consideration of correlation between asset classes) is more or less at the heart of modern portfolio theory.
Topic Author
aaronh
Posts: 17
Joined: Sat Feb 22, 2014 8:17 am

Re: Making a case for bonds (All About Asset Allocation)

Post by aaronh »

Rodc wrote:
To your first question: Yes rebalancing is market timing, but good market timing if one believes (which most of us do) that price movements in asset classes, like stocks and bonds, have reversion to mean, i.e. what goes up comes back down and vice versa.
Lots of very long threads show both statements are not universally held to be true. Indeed rebalancing as market timing seems to be a minority view. I won't rehash the arguments as there are already many hundreds of posts on the topic.

We have a current thread on reversion to the mean that is worth checking into as well. Very unclear that RTM is true, though it may depend on what one means by the term as there are at least a few definitions so one needs an actual precise definition.

The locked thread on why own bonds at all (http://www.bogleheads.org/forum/viewtop ... 0&t=132539) has some good arguments for and against holding bonds and there is no need to read past the first page or two to see them. This might be of some value to the OP.
Thanks for your thoughts. The locked thread that you reference is the thread that is causing me so much confusion as to what to do. I was convinced that I'd ditch my current all-equity portfolio and supplement it with bonds until I read that post. :(

However, nowhere in the thread (admittedly,I gave up trying to find after 4+ hours of reading), did anyone mention that a potential positive to owning bonds is that it can dampen the market highs and lows, which can yield higher compounded returns in the long run (the idea of making less (due to picking less volatile asset classes) can make you more in the long run due to compounding). The thread was 9 pages deep and I didn't want my question regarding this idea to get lost in a sea of bickering.
Topic Author
aaronh
Posts: 17
Joined: Sat Feb 22, 2014 8:17 am

Re: Making a case for bonds (All About Asset Allocation)

Post by aaronh »

abuss368 wrote:If the market decline and turned into a severe bear market for many years, could you stomach to see your equities cut in half or more?
Hi abuss368, are we talking bear market soon or bear market 5 years before retirement?
Topic Author
aaronh
Posts: 17
Joined: Sat Feb 22, 2014 8:17 am

Re: Making a case for bonds (All About Asset Allocation)

Post by aaronh »

placeholder wrote:Historically about the most consistent (at least in direction if not magnitude) negative correlator for stocks has been long bonds which you can run some Morningstar charts to see.
Thanks, this certainly helps me feel more optimistic in making a case for bonds in my portfolio.
Rodc
Posts: 13601
Joined: Tue Jun 26, 2007 9:46 am

Re: Making a case for bonds (All About Asset Allocation)

Post by Rodc »

However, nowhere in the thread (admittedly,I gave up trying to find after 4+ hours of reading), did anyone mention that a potential positive to owning bonds is that it can dampen the market highs and lows, which can yield higher compounded returns in the long run
Historically as shown in the graphs above (US data only, I did not run international), this seems to be the case only at the high bond end. That is some modest about of stocks added to a heavy bond portfolio, but not when adding a modest amount of bonds to a stock heavy portfolio.

While these are results I generated some time back, I have seen many similar results.

So if you are personally on the very high stock end, this may be more of a theoretical concept than what you will see.

That said, personally would err on the side of caution and hold some bonds. History may not repeat so I would not want to be 100% anything just on general principle.
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.
User avatar
abuss368
Posts: 27850
Joined: Mon Aug 03, 2009 2:33 pm
Location: Where the water is warm, the drinks are cold, and I don't know the names of the players!
Contact:

Re: Making a case for bonds (All About Asset Allocation)

Post by abuss368 »

aaronh wrote:
abuss368 wrote:If the market decline and turned into a severe bear market for many years, could you stomach to see your equities cut in half or more?
Hi abuss368, are we talking bear market soon or bear market 5 years before retirement?
From a risk stand point I am not so sure it matters. At the end of the day your investment strategy must pass the sleep test as Warren Buffett so often notes. If you are sleeping fine no problem.
John C. Bogle: “Simplicity is the master key to financial success."
Dandy
Posts: 6701
Joined: Sun Apr 25, 2010 7:42 pm

Re: Making a case for bonds (All About Asset Allocation)

Post by Dandy »

Dandy, According to R. Ferri in his Asset Allocation book, constant negatively correlated assets do not exist. So while you say "You are not predicting the future", doesn't it seem like we are predicting the future if we assume that stocks and bonds will have a negative correlation in the next 20+ years (my timeline)? I really want to understand, I feel as if there's a critical piece of information missing. Are we just saying that more times than not, there is a negative correlation between stocks and bonds?


First of all I think you are over thinking the issue. You won't get 100% agreement on any allocation. While bonds and stocks won't have negative correlations for every time period they do tend do well when held together. When both stocks and bonds decline usually bonds decline less. So, there is benefit even if the move in the same direction. I know that Mr Ferri suggests portfolios that have allocations to equities and bonds and suggests re balancing them.

Any decision about how to allocate your investments comes with some level of risk/concern. No matter how hard you search you won't find a totally comfortable decision or one that everyone's agrees is optimal. In my opinion most "experts" that I respect suggest holding at least some fixed income at almost any age. So please continue to try to understand investments but at some point you just have to pick an approach and stick to it.
good luck
Costanza
Posts: 52
Joined: Sun Dec 08, 2013 7:08 pm

Re: Making a case for bonds (All About Asset Allocation)

Post by Costanza »

Aaronh, you are looking for a silver bullet that doesn't exist. Add bonds only if you would prefer less volatility to maximum return. That is the one good reason to add bonds -- in my opinion, a good reason indeed. But there is no reason to expect higher returns from the addition of bonds. It could possibly happen, if stocks and bonds behave very differently from historical norms. Most likely, it won't.

True, lower volatility can increase returns, due to the difference between arithmetic and geometric mean. But don't draw the wrong implications. You have to apply this principle to a concrete situation. And when that situation is adding bonds to an all-stock portfolio, there is simply no good reason to project that bonds, with their lower expected return, will decrease volatility to such an extreme degree that overall returns will increase.
Sagenick48
Posts: 376
Joined: Sat Nov 02, 2013 8:22 am

Re: Making a case for bonds (All About Asset Allocation)

Post by Sagenick48 »

I will give you my thoughts from 40 plus years of investing. I am 66 and am probably 80% stocks and the rest cash and bonds. I started investing my 401k and Ira in the bear market of 1974+.

Bonds and cash are like anchors for a boat. They hold you in place in bad times and hold you back in good times.

The major benefit they provide is the emotional one of not being afraid when the inevitable crash occurs, whether it is the flash crash of 1987, the dot com crash of 2000, or the financial crisis of of 2008-9. Or the ups and downs in between.

So you can start at age 25 being 100% in growth stocks, switch to a more balanced portfolio as you age and then as you approach retirement can either take more risk again, because you have enough to weather any storm, or continue being conservative because you can't take as much risk. Finally when you get to 90 or so, assuming you still have your wits about you, you can go back to 100% growth stocks because your time horizon has become so long again, you are investing for your grandkids and great grandkids.
The market goes up, the market goes down.
User avatar
packer16
Posts: 1488
Joined: Sat Jan 04, 2014 1:28 pm

Re: Making a case for bonds (All About Asset Allocation)

Post by packer16 »

I agree that bonds provide more psychological safety than anything else (except of course for your cash needs). I think the key thing is you have to be honest with yourself. If having something going down 50% next year will bother you (I think most folks would say yes) then add some bonds. Some "different" folks (I am in that boat) it would not bother. Your financial circumstances can also play into this. I have a steady job and my biggest financial obligations will be my kids college which I should be able to fund with my salary and no savings (except 401(k))for the next 6 years.

As to having dry powder to deploy and outperforming a 100% equity portfolio, I think the odds are against that strategy.

Packer
Buy cheap and something good might happen
pradador
Posts: 156
Joined: Thu Jun 14, 2012 9:20 pm
Contact:

Re: Making a case for bonds (All About Asset Allocation)

Post by pradador »

Costanza wrote:Aaronh, you are looking for a silver bullet that doesn't exist. Add bonds only if you would prefer less volatility to maximum return. That is the one good reason to add bonds -- in my opinion, a good reason indeed. But there is no reason to expect higher returns from the addition of bonds. It could possibly happen, if stocks and bonds behave very differently from historical norms. Most likely, it won't.

True, lower volatility can increase returns, due to the difference between arithmetic and geometric mean. But don't draw the wrong implications. You have to apply this principle to a concrete situation. And when that situation is adding bonds to an all-stock portfolio, there is simply no good reason to project that bonds, with their lower expected return, will decrease volatility to such an extreme degree that overall returns will increase.
Would it not be better if everyone used annualized returns when comparing these numbers? There's some good data here: http://pages.stern.nyu.edu/~adamodar/Ne ... retSP.html

Code: Select all

Arithmetic Average 
          S&P 500  3-M Bill  10-Y Bond
1928-2013	11.50%	3.57%	5.21%
1964-2013	11.29%	5.11%	6.97%
2004-2013	9.10%	 1.56%	4.69%

Geometric Average				
1928-2013	9.55%	3.53%	4.93%
1964-2013	9.89%	5.07%	6.56%
2004-2013	7.34%	1.54%	4.27%
User avatar
Chan_va
Posts: 869
Joined: Wed Dec 05, 2012 6:15 pm

Re: Making a case for bonds (All About Asset Allocation)

Post by Chan_va »

Aaronh,

Implicit in your argument is the assumption that equities are safer the longer your time horizon. While that has been true in the past century, you need to ask yourself the fundamental question- why do equites have a higher expected return than bonds? The answer is simple and profound. It's because they are riskier.

So I own bonds in case the equity risk shows up in a big way. Even over the long term. Which is also why I rebalance from stocks to bonds but not vice versa. Like any insurance, owning bonds costs you if things go well. You will always have a higher expected return with 100% stocks than anything else. The equity risk premium is too high for any geometric mean stuff to matter.
cb474
Posts: 905
Joined: Tue Jan 19, 2010 5:32 am

Re: Making a case for bonds (All About Asset Allocation)

Post by cb474 »

dbr wrote:The definitive theory on the subject would be the work of Fama and French with the three factor model for investment returns (q.v.*). That research shows that returns of an asset class can be predicted to statistical accuracy that is high for financial work by regression on market risk, on a value factor (value stocks return more), and on a size factor (small stocks return more). The value metric in that theory is book to market ratio rather than P/E. It is a question of exploring around to find the greatest explanatory power. The general idea of the theory is to assume that return is related to risk although there is not a clearly understood rationale for why size and value are risks. In the theory it seems that one could say the definition of risk is "that which predicts return."
Thanks dbr.
packer16 wrote:Why wouldn't they be comparable? They are telling you how much you are paying for a dollar of earnings. I would want to pay less for smaller firms all things being equal. I think most of the time the SCV does have a lower P/E than the LCV as a premium is being paid for stability and liquidity. Given the parity, I would probably have less tilt than I would if SCV has a P/E less than LCV.
Well, my question was why would it (i.e. why should I assume this)? So "why wouldn't it?" isn't much of an answer. Notably, as dbr explained, for value funds, it turns out book to market ratio has more explanatory relevance for future returns, than P/E. As you yourself note, small firms have different risks that larger ones, making earnings not the only relevant metric and perhaps not relevant in the same way or to the same degree. Also, people talk all the time about the historical average P/E for the S&P, but is it the same for other indexes? I don't know. I'm having trouble even finding that type of information. Ultimately, I've learned that it is pretty much always a mistake to assume that a concept applied to one thing, applies in the same way with all the same signficance to something else that superficially looks similar. It's an easy and appealing way to think, but more often than not assumes all sorts of false equivalencies. So that's why I asked. I just haven't seen much discussion (especially historically) of P/E beyond how it applies to the S&P.
User avatar
packer16
Posts: 1488
Joined: Sat Jan 04, 2014 1:28 pm

Re: Making a case for bonds (All About Asset Allocation)

Post by packer16 »

I think the general idea of finding out how cheap something is (especially if that is what you are trying to get exposure to) is important. My favorite multiples are P/FCF and EV/EBITDA but you get the similar result using different metrics. The issue with P/BV is that post 2000 there was a massive accounting change which stated all acquisitions must be asset acquisition. Before that it was a choice. With such a large change in the definition of underlying metric, the comparability between the pre and post periods becomes an issue (It is an issue with reported earnings also). With FCF and EBITDA the issue goes away due to definition being the same throughout history.

My observation is the SCV funds have drifted to not so small cap (average market cap in the billions) and not so much value (multiples in the mid to high teens) with higher than average value metrics. I guess what I am suggesting is some absolute metrics for size and value in addition to relative metrics. So you may to focus on large cap value as you are getting about the same exposure to value with much more stable and liquid firms.

Packer
Buy cheap and something good might happen
Post Reply