Explain the 10% in Buffet's 90/10 Portfolio

[heybro]

Regarding Warren Buffet's 90/10 Portfolio, he says the 10% is US Treasuries. Can anyone explain this 10% more. Is he buying individual Treasury Bonds? How often is he renewing them? How does it work in a retirement account in terms of purchasing and repurchasing?  Do you set up a ladder much like a CD ladder?

If a person was going to hold 10% in this type of asset class, what would be the pros to buying Treasury Bonds vs. a different type of bond or bond fund or simple money market fund for that matter.

I'm not trying to start any age old debates about what asset allocation is better but rather asking about the explanation for this particular asset allocation and how this particular one works.

[Radagast]

Regarding Warren Buffet's 90/10 Portfolio, he says the 10% is US Treasuries. Can anyone explain this 10% more. Is he buying individual Treasury Bonds? How often is he renewing them? How does it work in a retirement account in terms of purchasing and repurchasing?  Do you set up a ladder much like a CD ladder?

If a person was going to hold 10% in this type of asset class, what would be the pros to buying Treasury Bonds vs. a different type of bond or bond fund or simple money market fund for that matter.

I'm not trying to start any age old debates about what asset allocation is better but rather asking about the explanation for this particular asset allocation and how this particular one works.

I understand Berkshire dominates the global market for short term treasury bills. He renews constantly. You can do with a money market or short term bond fund. I think I read some brokers allow you to set up an automatically rolling treasury ladder for free. I believe Treasury Direct also allow allows this. More research needed.

Treasury bonds are safer, and in theory buying them directly eliminates counter party risk. As a small fish FDIC/NCUA insured CD's, saving accounts, and (Treasury) I-bonds often have better yields for similar or the same risk. The downside is these things are very stable and there will be little or no "flight to safety" effect, which if it occurs means longer term bond prices might rise in response to a crisis, but these won't.

The biggest decision for high quality short or intermediate term bonds and bond-like-things is "how much." "What type" is not nearly as important as long as you are keeping your yields somewhat near the market yield (no 0% Wells Fargo cash or junk bonds in this category).

[neil]

Once coming to the conclusion that your AA will be 90% standard market index risk/return, 10% low fee low risk fixed income, the main idea here is that the 10% is not really for "return" and we want to minimize risk as much as possible.  Functionally, I'm not sure you can get lower risk than buying short term T-bills directly from the agent itself which effectively cannot default (save for some systemic collapse that probably screws everything up).  You can sit there and try to play games with duration, or quality, or whatever, but the stocks are where you take the risk and get the return and that is not really what the bonds are for.  Market timing in the bond market is no better than the stock market.

I don't know if he ever went into detail, but I would say it doesn't matter because I don't think how we execute our personal finance necessarily matches his family.  If I were to do this, I'd probably ladder based on a monthly expense number and break up my allocation that way.  I guess I'd start buying 2yr bonds around $4-5K each, renew them as they mature and try to live off the stocks.  Then, when stocks take a hit I'd withdraw the bonds instead and repair the ladder on a recovery.

This seems like a lot of micromanagement at these numbers.  With eight or nine figures, it probably wouldn't be as fiddly, and maybe worth paying someone to do it (rather than paying the ~.1% fee).  VFIRX is probably fine for most of us.  Duration 2 years vs 6 for VBTLX.  CDs or online savings is also fine.  It's probably not theoretically the lowest risk, but low enough without complicating my life.

[heybro]

Once coming to the conclusion that your AA will be 90% standard market index risk/return, 10% low fee low risk fixed income, the main idea here is that the 10% is not really for "return" and we want to minimize risk as much as possible.  Functionally, I'm not sure you can get lower risk than buying short term T-bills directly from the agent itself which effectively cannot default.

This is what I was thinking was the theory behind this.  All your risk is in the 90 percent stocks so you go to the safest thing possible for the 10 percent bonds: Treasury Bills.  So, if this is the theory, what would a Total Bond Market fund or any other type of bond mean?  They would be more risky and not as simplified?  I just want to be sure that if the theory is 'all risk in the 90 percent and all safety in the 10 percent' that it is indeed Treasury Bills that answer that.  Would that be a short term treasury fund?  What about intermediate or long term treasury fund?  Or is the difference pretty mute as Radagast said.

[Telecaster]

This is what I was thinking was the theory behind this.  All your risk is in the 90 percent stocks so you go to the safest thing possible for the 10 percent bonds: Treasury Bills.  So, if this is the theory, what would a Total Bond Market fund or any other type of bond mean?  They would be more risky and not as simplified?  I just want to be sure that if the theory is 'all risk in the 90 percent and all safety in the 10 percent' that it is indeed Treasury Bills that answer that.  Would that be a short term treasury fund?  What about intermediate or long term treasury fund?  Or is the difference pretty mute as Radagast said.

Longer term bonds have inflation risk.   The point is you want some stability in your portfolio.  Stirring in some bonds actually improves returns over a 100% stock portfolio. 

[neil]

This is a pretty nice visualization:

http://stockcharts.com/freecharts/yieldcurve.php

When you buy VBTLX, basically you own the entire yield curve.  (But also corporate bons, MBS, etc, adding a range of risk as well.)  If you turn on animation, you can see the curve go up and down.  Of course, if yields go up, bond values fall, but time makes up for this through higher coupon rate.  (And vice versa).  You'll notice that it's common for the entire yield curve to go up or down together, but it's definitely possible and normal for different parts of the curve to move differently.

When a mutual fund focuses on short, medium or long term, it means the fund is focused on that part of the yield curve, rather than a mix of everything.  But you also must take into account that the longer the term, the more your bond's secondary value changes with the change in interest rate, because maturity is much farther away.  (You can use a bond calculator to see how a bond's value is affected by change in rates.)

By switching VBTLX for VFIRX in your portfolio, you're probably saying a few things:
- I don't need corp bonds (probably because 90% of your portfolio is exposed to corp upside and risk already)
- I don't want interest rate risk
- Short term bonds are normally close enough to cover inflation
- I don't care if I miss out on bond upside
- Diversification is not relevant because the government cannot default (*in any meaningful way that can be protected against though an alternative)

I think the points are valid and interesting to discuss.  I am also not smart enough to know that one is better than another.  Given the scenario where you're talking about a 10% allocation and few basis points of real return, it probably doesn't matter too much.  As a retail investor, it is as easy to buy VBTLX as it is to buy VFIRX.

I wouldn't say VBTLX is bad just because it carries some additional risk.  I plan to hold some over my entire lifetime (which hopefully is a long time) and I will essentially see every bond issue mature and be replaced in that time.  And some minority of corp bonds will probably default.  On average, I'll probably see a real return of about 1% or so.  As I withdraw, most of the time I'll be selling stocks anyway because they should have a higher rate of return anyway.  The bonds are there as ballast (as JL Collins puts it) and usually have strong returns when stocks aren't.

http://jlcollinsnh.com/2017/12/20/the-bond-experiment-return-to-vbtlx/

Personally, I think the discussion is interesting and I enjoy learning more, but in the end this is relatively small portion of your portfolio.  It is probably more important to be comfortable with your selection than deciphering what is optimal.

[Jamese20]

This is what I was thinking was the theory behind this.  All your risk is in the 90 percent stocks so you go to the safest thing possible for the 10 percent bonds: Treasury Bills.  So, if this is the theory, what would a Total Bond Market fund or any other type of bond mean?  They would be more risky and not as simplified?  I just want to be sure that if the theory is 'all risk in the 90 percent and all safety in the 10 percent' that it is indeed Treasury Bills that answer that.  Would that be a short term treasury fund?  What about intermediate or long term treasury fund?  Or is the difference pretty mute as Radagast said.

Longer term bonds have inflation risk.   The point is you want some stability in your portfolio.  Stirring in some bonds actually improves returns over a 100% stock portfolio.

Can you link the source to this statement please, I have yet to see any mix that provides more returns than 100 % stock