Short, Medium & Long Term Bond Funds 2013

[aaronc879]

Quote:

Originally Posted by haha An hour of basic reading on the internet will explain these things for you far better than a discussion here. There are people here who understand these things very well, but they are not guaranteed to be the only ones who might post about them. It can seem confusing. Ha

I've spent the last couple days reading a couple dozen articles online and all it got me was confused. I guess i'll just hold what I have.

[NW-Bound]

It depends on how much the yield of that bond rises or falls.

As a refresher (for myself too), the present value of a bond is the sum of the discounted value of its income stream and the discounted value of its principal that is paid back at maturity.

Here's an example. For a 10-yr bond of $1000 principal paying 3% interest a year ($30/yr), if the current interest rate (or inflation?) is 0%, then its value is $1000 + 10 * $30 = $1300. If the current rate is 3%, then the discounted value of the interest payments is $255.906 and that of the principal is $744.09, and the sum is exactly $1000. The first case is trivial and can be easily seen. The latter case is more complicated, yet the result is as can be expected, because the interest payment exactly cancels out the inflation.

The actual formulas are PVA = I[1-(1+k)^-n]/k and PV = FV/(1+k)^n, where PVA and PV are the values of the interest stream and the principal, FV is the returned principal at maturity, I is the annual interest payment, k is the current interest rate (inflation rate?), and n is the number of annual payments. Note that when k=0, then the formula blows up but we already know that PVA = I * n, and PV = FV for that limiting trivial case. Or if one wants to be fancy, remind yourself of freshman calculus and take the limit when k approaches 0.

The problem is the new interest rate is assumed constant in the remaining duration of the bond, which we know is not true. And the bond yields of different duration often have a strenuous relationship. Long bonds should have a higher coupon rate than short-duration bonds, but the yield curve has been inverted in the past.

And in the above formula, should I use expected inflation rate which reflects the discounted value of the principal and interest stream, or do I use the current bond yield which shows the lost opportunity relative to selling this bond and buying a new one now?

It's really weird stuff, and I admit that I have not been following this closely in the past. I made most of my money from stocks, but now that I am not working and have more time when not RV'ing, I am willing to learn a new trick.

However, I will note that if the bond yield changes 2% from a high value like 6% to 8%, the effect is nowhere as big as a change from a low value of 2% to 4%, which we observed recently.

"In theory there is no difference between theory and practice. In practice there is." -- YogiBerra

[haha]

Quote:

Originally Posted by aaronc879 I've spent the last couple days reading a couple dozen articles online and all it got me was confused. I guess i'll just hold what I have.

I have a very tenuous understanding of this, not enough that I want to try to teach others.

Here is a decent non-technical article. Bond investors face a reckoning as interest rates jump - Page 2 - Los Angeles Times

Also, I have a bond fund that was running a duration of around 4 years last spring when 10 year interest rates abruptly rose, as I remember from about 1.7% to almost 3%. My fund dropped roughly 5% in quoted value, which is close to the short cut prediction of capital value changes = duration * delta interest rates. I am using 10 year rates, because they are easily found, and because this fund owned a lot of mbs, and rates on these tend to key off 10 year government rates. Of course if you are re-investing dividends, your share count starts rising and tending to close the gap created by that loss. I have noticed that it does not seem to pay to be extremely conservative regarding duration, because a higher interest rate combined with dividend reinvestment can catch up to cash or near cash pretty quickly. And of course if rates fall, you are much better off. I do not subscribe to the meme that rates from here must necessarily change only in one direction-up. If 10 year rates were 1.7% last April, is it impossible for them to be 1.7% once more? If the answer is yes, why would this be true?

Maybe some expert will have time to stop by later.

Ha

[FIRE'd@51]

Quote:

Originally Posted by aaronc879 I have a fairly large portion of my portfolio in a rollover IRA invested in PGBOX. Is that an example of a "bad" bond fund when rates go up? I put the money there because I thought it was far more conservative than anything involving equities. Now i'm not so sure.

A problem I see with PGBOX is its expense ratio of 0.74%. Vanguard probably has a very similar fund with an expense ratio of 0.10%.

[nun]

Quote:

Originally Posted by veremchuka I'm 50% in FI. Vanguard short term investment grade and intermediate term investment grade and stable value in my 401k. The ITIG fund makes me nervous, the STIG not so much. Since I reinvest dividends in both and have no distributions planned until 70 1/2, the ITIG fund's duration may not be a problem as that's a bit beyond the 5-6 years of the ITIG fund and RMD.

That's my plan too.

My only bonds are those I hold in Wellesley and 7% of my AA in Vanguard Intermediate Investment Grade. The majority of my fixed income is in cash, stable value and TIAA-traditional. Last year I moved all my TBM allocation to Stable Value in my 457 to prepare for ER so that I have enough to cover 7 years of spending from 52.5 to 59.5. I don't plan to touch the bonds I still have for at least 10 years.

[pb4uski]

Quote:

Originally Posted by aaronc879 I've spent the last couple days reading a couple dozen articles online and all it got me was confused. I guess i'll just hold what I have.

For each 1% rise in interest rates a bond (or bond funds) value would be expected to decrease by its duration divided by 100.

So if the duration is 5.4 and rates rise 1% then the bond's value would be expected to decline by 5.4%. If rates decline 0.5% then the bond's value would be expected to increase by 2.7%.

[racy]

Quote:

Originally Posted by aaronc879 I've spent the last couple days reading a couple dozen articles online and all it got me was confused. I guess i'll just hold what I have.

Hi aaron,
Here's something you might find interesting on Vanguard's site:
https://investor.vanguard.com/insigh...asics-duration

[FIRE'd@51]

Quote:

Originally Posted by pb4uski For each 1% rise in interest rates a bond (or bond funds) value would be expected to decrease by its duration divided by 100. So if the duration is 5.4 and rates rise 1% then the bond's value would be expected to decline by 5.4%. If rates decline 0.5% then the bond's value would be expected to increase by 2.7%.

This is a good rule of thumb for relatively small changes in interest rates, but one should be aware that it overestimates (underestimates) the price change for large moves up (down) in rates, because the duration itself is a function of the interest rate.

[aaronc879]

Quote:

Originally Posted by pb4uski For each 1% rise in interest rates a bond (or bond funds) value would be expected to decrease by its duration divided by 100. So if the duration is 5.4 and rates rise 1% then the bond's value would be expected to decline by 5.4%. If rates decline 0.5% then the bond's value would be expected to increase by 2.7%.

Thanks. That helps a lot. I read an article that said the Fed doesn't plan to raise rates until unemployment gets down to 6.5%. That won't happen in 2014 most likely so bonds should still be ok this year, right?

[pb4uski]

Quote:

Originally Posted by FIRE'd@51 This is a good rule of thumb for relatively small changes in interest rates, but one should be aware that it overestimates (underestimates) the price change for large moves up (down) in rates, because the duration itself is a function of the interest rate.

Agreed that the general rule decays some with very significant changes in rates.

[pb4uski]

Quote:

Originally Posted by aaronc879 ...That won't happen in 2014 most likely so bonds should still be ok this year, right?

In need to think about that one for a while. Get back to me in December and I'll have a solid answer for you.

[FIRE'd@51]

Quote:

Originally Posted by aaronc879 I read an article that said the Fed doesn't plan to raise rates until unemployment gets down to 6.5%. That won't happen in 2014 most likely so bonds should still be ok this year, right?

Not necessarily. The Fed has direct control over very short-term rates (i.e. the overnight rate), but its control over intermediate and long-term rates is much more tenuous.

[racy]

Quote:

Originally Posted by Midpack Hopefully there's no one left here who doesn't recognize the risk associated with long duration bond funds when interest rates eventually increase. ...

Out of curiosity I looked to see what Vanguard does with their Target Retirement Income Fund, "designed for investors already in retirement". It holds approximately 70% in bonds. The bonds, percentage and duration are:

Total Bond Fund, 39.3%, 5.5 yrs
S-T Inflation Protected, 16.8%, 2.4 yrs
Total Int'l Bond Index, 14.0%, 6.6 yrs.

So, the weighted duration is 5.5 yrs, an intermediate duration fund.

[veremchuka]

Quote:

Originally Posted by aaronc879 So what are we taking about here? When rates rise, an intermediate-term fund will fall how far? 5%/yr,10%/yr? Even in bad years a bond fund won't fall 20%+ like equities, right?

There have been several replies but how about this -

say you own a long term bond fund (a very bad idea now), the duration is 14 years, say interest rates rise 2% over 2 years. You just lost 28% of the fund's nav! So it all depends upon the DURATION of the fund. If you don't know the duration of a bond fund you are in then you better find out because it is important in a rising rate environment.

[aaronc879]

Quote:

Originally Posted by veremchuka There have been several replies but how about this - say you own a long term bond fund (a very bad idea now), the duration is 14 years, say interest rates rise 2% over 2 years. You just lost 28% of the fund's nav! So it all depends upon the DURATION of the fund. If you don't know the duration of a bond fund you are in then you better find out because it is important in a rising rate environment.

Do you know what duration the bonds are in Wellesley?

[aaronc879]

Quote:

Originally Posted by aaronc879 Do you know what duration the bonds are in Wellesley?

Nevermind, I found it. Moderate-term investment grade.

[Lsbcal]

This M* interview with Bogle on fixing the Total Bond Mkt Fund and what to do now is worth listening to:
Bogle: We Need to Fix the Bond Index

[clifp]

Quote:

Originally Posted by Lsbcal This M* interview with Bogle on fixing the Total Bond Mkt Fund and what to do now is worth listening to: Bogle: We Need to Fix the Bond Index

I think this link is better. I think any folks with large positions in the total bond fund or BND should listen to this interview.

I made a similar point in different thread, not knowing that Bogle himself agreed with me..

[veremchuka]

I'm not the 1st to point this out but the Total Bond Market Index Fund is far from being "total". There's no high yield or international or emerging markets bonds. The GNMA and treasuries in the TBMI is killing the yield.

[haha]

Quote:

Originally Posted by clifp I think this link is better. I think any folks with large positions in the total bond fund or BND should listen to this interview. I made a similar point in different thread, not knowing that Bogle himself agreed with me..

Maybe what he says is still relevant, but there have been huge changes in interest rates since this interview on April 17, 2013. Bogle refers to a couple of treasuries, like the 10 year, which at that time was well under 2% (around 1.7%), and now is just slightly below 3%.

Ha