Author Topic: Total US Treasury Fund vs. Short, Intermediate, & Long  (Read 606 times)

heybro

  • Stubble
  • **
  • Posts: 198
Total US Treasury Fund vs. Short, Intermediate, & Long
« on: September 25, 2018, 03:48:04 PM »
I've got a 100% Equity portfolio and am trying to figure out my eventual strategy in to bonds.  I prefer US Treasuries.

How do you like the GOVT iShares U.S. Treasury Bond ETF which includes a mix of short, intermediate, and long term US Treasuries.

The other option is to stress over picking between a short term fund, intermediate fund, or long fund, which no one seems to have a clear answer on other than 'interest rates will rise so stick with short,' but if I do that, how do I goal post when to start buying intermediate and long, etc. if ever.

DavidAnnArbor

  • Handlebar Stache
  • *****
  • Posts: 2138
  • Age: 54
  • Location: Ann Arbor, Michigan
Re: Total US Treasury Fund vs. Short, Intermediate, & Long
« Reply #1 on: September 25, 2018, 08:23:49 PM »
It's hard to predict the future so generally you can choose any bond index fund that is well diversified 

Radagast

  • Handlebar Stache
  • *****
  • Posts: 1589
  • Location: West of the Mountains, East of the Sea
  • One does not simply work into Mordor
Re: Total US Treasury Fund vs. Short, Intermediate, & Long
« Reply #2 on: September 25, 2018, 08:28:18 PM »
There is not exactly a right answer, any of these are correct depending on your goal.
-Short term government bonds: the classic low risk low return investment. Supposedly these are least correlated to the stock market, but their movement is so small it doesn't really matter. Fighting fire (stock market risk) with water (low risk).
-Intermediate term government bonds: These have (obviously) an intermediate amount of duration risk. However, in a normally sloping yield curve you get an additional capital return because they generally buy long term bonds at disproportionately low prices (high rates) and then sell at high prices (lower rates). That often gives them an extra amount of return per unit of risk, which should almost always be the case for fiat currencies (under the gold standard the yield curve was commonly inverted by modern standards, as investors sought long term predictability). Some people try to buy bonds or funds to target the steepest part of the yield curve to hopefully maximize return per unit risk.
-Long term government bonds: These have a large amount of duration risk, and little additional capital gain from the shape of the yield curve. Their price generally moves around with similar volatility to the stock market, but generally with poor correlation. These are fighting fire with fire: when they work they will produce amazing results, but if the wind shifts they can burn you. If you have 20% or less allocated to bonds, there is evidence that long term bonds will give you the best result more often than not, but on rare occasions they will be really bad at the wrong time.
-"Total" government bond: This is nice as a theory, but in practice it gives you similar risk to intermediate term bonds, while providing little additional capital gain from the shape of the yield curve. They usually have similar risk and lower returns to intermediate bond funds. I decided I am not a big fan even though the "don't outguess the market" advice should apply, but I guess that in this case it systematically does not apply. If it ever does, you will know because the yield curve will be flat, or inverted, or uniformly sloped for decades at a time. If none of those is the case, then an intermediate fund is better.
-Barbell: mix of short and long term bonds, skipping the middle. Also commonly advocated. Allows you to take a lot of interest rate risk for the occasional blow outs, while still having a safe cushion for the opposite blowouts. I like the idea of a mix of I-bonds and EDV for a barbell. I like I-bonds because they have OK returns and no risk at all (or as close as humanly possible) because they are a direct interaction between you and the government, so the market and other players cannot affect the price you get for them.
Another common suggestion is that the duration of your bond allocation should be similar to the time until you need money.

Anybody who says they know where interest rates are headed should provide a maturity, a date, and an approximate rate for their prediction, ie "one year from now 10 year treasury bonds will be 3.5-4.0%" so we can laugh at them one year from now.