Bonds - Short Term or Longer

[karluk]

Quote:

Originally Posted by mathjak107 you are way out in left field. if the intermediate bond rate was 2.06 when you bought in as you are hypothetically saying and rose to 5.62% then if you stayed for the duration figure of the fund currently 5.06 years you would simply get around the 2.06% rate you origonally got the day you bought in. that is what the duration of a fund means. it means for every point rates go up if you stay the duration you will get whole again back to the number you had the day you bought in. however any downgrades in credit throw that number out the window in a fund like total bond and you may never get back to that 2.06%. if rates are going higher and higher over those 4 years i doubt many would just sit and get 2.06% when rates are 3,4 and eventually 5% elsewherel you could but not many will. last year when nbonds took a hit did you see the billions that fled the bond funds ?

I guess my basic reaction is that for someone whose handle is "mathjak" you do an awfully lot of hand waving and hardly any concrete mathematical analysis. The math favors the intermediate bond fund, so vague claims to the contrary don't carry a lot of weight.

As far as investors real life behavior, I can't speak for others, but I definitely can speak for myself. I was a big net buyer of bond funds in 2013, when interest rates were rising, and a big net seller in 2014 and January of this year, when interest rates were falling. This was all in the name of rebalancing, but I know I made a substantial profit on these transactions over and above a simple buy and hold strategy. To me this is a huge advantage of the intermediate term bond funds on top of their in-built advantage over cash. It's possible to realistically implement a "buy low, sell high" strategy that isn't available with cash and CD investments.

[mathjak107]

believe what you want but rember i did try to correct your logic which is wrong. watch what happens to total bond fund if bond rates take a 3% jump

i am done .

[audreyh1]

Quote:

Originally Posted by mathjak107 believe what you want but rember i did try to correct your logic which is wrong. watch what happens to total bond fund if bond rates take a 3% jump i am done .

Exactly which rate do you think is going to jump 3%??

You can't lump all interest rates together. There is the Fed Funds rate, and then there are the rates for each duration of Treasury, then there are rates for various durations of different bond asset classes - muni, corporate, high yield etc.

IMO it's useless to paint them with a broad brush. The rates at different durations and among different bond asset classes operate quite differently.

[mathjak107]

When discussing bond funds the rate that matters is the rates that apply. Generally the 10 year treasury is the bench mark for intermediate term bonds.

What has a corrolation no better than a coin toss to intermediate bonds or long term bonds are the feds action on short term rates.

the chart posted above shows how a big 3.08% move over 4 years didn't effect intermediate bond funds like total bond much . but that chart fails to point out the 5 year bond was reflecting the feds short term moves which were huge.

the fact is the 10 year treasury went only from 4.06 to i think it was 4.76%. that is hardly an abnormal move in any year and over 4 years it is nothing to speak of and is quite tame so of course total bond reflects little change . see what happens if the 10 year increases over 4 years by 3.08% to a fund like total bond or corporate bond. you can look at the funds duration number and do your own math. .

all bonds are moved by rate changes and the catalyst is usually equivelant maturity treasury bonds. corporates ,muni's and high yield just add other parameters in such as credit rating. interest rates can stay the same but a rash of credit down grades will see corporate bonds fall.

we saw that in 2008-2009 when treasury bonds soared and corporates ranged from down 8-9% to up slightly.

[audreyh1]

Quote:

Originally Posted by mathjak107 When discussing bond funds the rate that matters is the rates that apply. Generally the 10 year treasury is the bench mark for intermediate term bonds. What has a corrolation no better than a coin toss to intermediate term and longer are the feds action on short term rates.

The 5 year treasury would be the appropriate rate to track for intermediate bond funds. Intermediate bond funds are usually around 5 years in duration. 5 year interest rates and 10 year interest rates can behave differently, and often do.

[mathjak107]

Quote:

Originally Posted by audreyh1 The 5 year treasury would be the appropriate rate to track for intermediate bond funds. Intermediate bond funds are usually around 5 years in duration. 5 year interest rates and 10 year interest rates can behave differently, and often do.

Just because a bond fund has an "average duration" of 5 years does not mean all the bonds in the fund have a duration of 5 years - a portion of the fund could have much longer duration. Take for instance the Intermediate Bond Category index -- the category average is that these bonds have about 20% in the 20-30 year bond category.

maybe the rest are in the 10 year and 3 year.

it is the weighted average of all these bonds that alone move to different bond rates that produce a duration figure and the funds return..

if you had 30 year bonds , some 15 year and some 3 year depending on how much of each you have you could have a 5 year average but in no way would your mix track a 5 year bond.

some will move with the 30 some with the 15 and some with the 3 year note. a 5 year bonds movement may have very little correlation to your actual moves since they all move differently.

but if whatever rates changed enough to move your fund 3% in 4 years will have a big effect on your total returns.

don't forget once you get away from treasuries greed ,fear and perception play a part too in a bonds price.

[audreyh1]

Quote:

Originally Posted by mathjak107 Just because a bond fund has an "average duration" of 5 years does not mean all the bonds in the fund have a duration of 5 years - a portion of the fund could have much longer duration. Take for instance the Intermediate Bond Category index -- the category average is that these bonds have about 20% in the 20-30 year bond category. maybe the rest are in the 10 year and 3 year. it is the weighted average of all these bonds that alone move to different bond rates that produce a duration figure and the funds return.. if you had 30 year bonds , some 15 year and some 3 year depending on how much of each you have you could have a 5 year average but in no way would your mix track a 5 year bond. some will move with the 30 some with the 15 and some with the 3 year note. a 5 year bonds movement may have very little correlation to your actuual moves since they all move differently.

And yet the 10 year Treasury is somehow magically the benchmark?

There is a reason that average duration is useful for modeling/predictive purposes.

Oh, never mind!

[mathjak107]

no the benchmark is not the 10 year . think of a bond fund that had equal weighting in 30 year and 1 year paper. hypothetically they may have a duration of 15 or so but they will track a 30 year bond and a 1 year note.. they would have zero to do with tracking a 15 year bond. the 15 year is far to long to track what a 1 year note is doing to even make sense and far to short to track a 30 year bond and what investors are asking for rates to go out that far.

now you should follow why i said the 5 year bond was the wrong benchmark to try to track an intermediate term bond fund with bonds ranging from maturing tuesday to 20 years out.. the weighting of the fund has nothing to do with the benchmarks that trade daily and set the course of each bonds rates.

duration is only the average weighting of the funds effect on the funds value when rates rise or fall by 1%.
duration numbers say hypothetically we can expect the funds value to fall 15% if overall the funds uinderlying bonds move 1%.

perhaps the 1 year moved zero but the 30 year moved 2%. so the funds weighting produced a 15% loss. that is how duration works.

now as far as the amount of years you have to stay that is a different issue. in this case eventually as those 1 year bonds roll over we will get new bonds paying 1% more but we have a crap load of 30 year bonds that are stuck at what they are. getting . the funds value has to drop 15% so anyone buying in now gets the higher rate even on the 30 year bonds which are actually earning the old lower interest rate before newer bonds paid 1% more .

after 15 years collecting that extra 1 % we got back the 15% the fund fell and we are whole again back to the interest rate deal we had the day we bought in.

from this point on you can pocket a little extra interest going forward but that assumes rates didn't move another point up in which case the process starts all over.

that is where the years fit in.

it is a bit trickier in real life since bond funds buy and sell bonds all day trying to capture small differences between years or types but like night follows day you can bet you will fall close to the funds duration number.

[karluk]

There are numerous inaccuracies and oversimplifications in Mathjak's explanation of duration. But this is a very complicated subject, so I'll content myself with providing a link that explains some of the basics:

Advanced Bond Concepts: Duration | Investopedia


From the point of view of the current thread, which was started to discuss whether short or intermediate term bonds were preferable, the paragraph on "immunization" is worth looking at. Quoting from the link

Quote:

Immunization As we mentioned in the above section, the interrelated factors of duration, coupon rate, term to maturity and price volatility are important for those investors employing duration-based immunization strategies. These strategies aim to match the durations of assets and liabilities within a portfolio for the purpose of minimizing the impact of interest rates on the net worth. To create these strategies, portfolio managers use Macaulay duration.

In my view, this concept of immunization explains quite clearly why so many investors go seriously awry with their bond allocations. They are spending too much time looking into their (rather cloudy) crystal balls trying to determine the future of interest rates, when they should actually be focusing on matching their bond mix with their future financial needs. Failing to do so will work extremely well if you're lucky enough to make good guesses about interest rates, but it can be a true disaster if you're wrong. Pick maturities that are too short and you risk reinvesting money that you won't need for 10-20 years at near zero interest rates. That must be a common regret among investors since 2007. Pick maturities that are too long and you're vulnerable to principal losses when you really need the money to be there. That could easily be a problem in the future.

So the real answer to OP's question is, "It depends." In an ideal world you would structure your fixed income investments to match your expected financial needs in order to protect yourself against interest rates not behaving the way you expect. Unfortunately, it's not always possible to know one's spending needs so many years in advance, which is probably why investors skew towards short term investments, such as setting up an emergency fund that is much bigger than they most likely will need.

[OrcasIslandBound]

Yes, bonds are actually a bit trickier to manage then stocks. Lots of good, heated discussion points by passionate people.

My own investment strategy actually uses more long term, 14 year duration, bonds for about a third of all the bonds that I have. One of the best diversifying assets for stocks is long term treasuries. This is from "The Bond Book" by Annette Thau. Will I take a hit in my portfolio because of this decision? Could happen. Did happen to someone in 2013 using the exact same tool I use, Vanguard long term bond index, BLV. Now, when I bought these bonds in September last year, the yield rate was about 4.5%. I'm aiming to retire in 2 years and that yield will hold up in payments for the duration of 14 years and at that time I will roughly get my full principal back as well.

But I don't intend to sell it then. In fact, if the interest rates do rise, I'll also likely have more gains in stocks and will simply buy even more BLV. I expect a much lower volatility with the overall 43% stocks and 57% bonds portfolio because of the long term treasuries component.

Sent from my Nexus 4 using Early Retirement Forum mobile app

[mathjak107]

oh it is accurate enough and simplified enough to understand why the info above was wrong.

we have never had an extended bear market in intermediate term bonds yet since 1980.

there is no comparison or data you can use to see what if 's. we had single years where they lost money but we have had zero extended bear markets in bonds in 35 years to compare against.

in fact in every year the fed raised short term rates 1% or more intermediate term bond funds ignored the short term rise and produced positive returns except 1 year 1994 . in that year investors in bonds agreed with the feds acrtion and bid longer term bond rates higher as well so many funds lost money and posted negative returns .

the barclay intermediate term index posted negative returns of 2% after figuring in all the interest .

the issues will come when bond investors want more compensation and bid these longer rates higher. if you think that is a good thing for your existing bond investment you are dreaming and need a math lesson..

[HF63]

I decided to keep the money in cash since the dollar keeps rising in value versus the Mexican peso. The dollar is up to almost 15 pesos and this has provided a nice return with no taxes!

[karluk]

Quote:

Originally Posted by mathjak107 oh it is accurate enough and simplified enough to understand why the info above was wrong.

No, it is inaccurate enough and oversimplified enought to lead you to erroneous conclusions. You have indulged in a lot of hand waving in this thread in order to come to faulty conclusions about how bonds behave, but I am a numbers guy and ran some typical rising interest rate scenarios yesterday to convince myself that bond fund investors do, indeed profit from rising interest rates. Over time the decline in NAV from the rate increase is fully compensated by reinvesting dividends at a higher yield. The payback period varies depending on the assumptions one makes about the size and frequency of the rate increases, but is highly dependent on duration, as can be expected from basic mathematical analysis.

So, if you want to make a positive contribution to this thread, you should seriously rethink your unwillingness to work through the math and present us with your prediction of how interest rates will change in the future along with some concrete calculations (using your prediction) to determine how long it will take for holders of intermediate bonds to be better off than holders of short term t-bills, or whatever your favorite cash equivalent is. If you think you can present a plausible scenario where cash is permanently better than intermediate term bonds, that's fine. We can critique your calculations and either agree with you or show you where your calculations are wrong.

What isn't productive is all of this hand waving that allows you to present unsupported assertions as fact, and substitute your opinion about how bonds should behave with how they actually do behave.

[mathjak107]

time will prove who is right and who is wrong!

fully compensated means you worked your way back over the duration period to the deal you had day one that you bought in. that is likely below what the current interest rate would have payed in shorter term instruments over the same time frame.

the greater the rise the more left behind current rates you will be.

eventually you will be whole again but that is only true if rates stop rising. with historial rates on bonds in the 6-7% range and us at 2% you can be behind the curve for a very long time if rates rise for a very long time.

in a corporate bond fund if credt ratings are cut the duration period can grow longer and longer as nav falls not only from rates rising but bond values falling from down grades like we saw in 2008-2009 and all bets are off as to when that compensated point will be hit.

remember too that funds can shorten durations when rates rise meaning when that 10 year bond matures they buy a 5 year instead . that means lower interest rates than 10 year bonds would get so you can't count to much on those duration numbers making you whole in a specified time period.

my opinion is once bond rates start to rise one would be foolish to sit static in conventional bonds and would do much better in other types of income funds more appropriate.

you wan't to ? be my guest . but it isn't something i would do . i would keep some money in coventional bonds but certainly not 40% of the bond budget the way i would have when rates were falling for 35 years,..

[pb4uski]

Quote:

Originally Posted by mathjak107 ....i am done .

When is done, done?

[Mulligan]

Quote:

Originally Posted by pb4uski When is done, done?

Hey now PB, I'm enjoying Karluk's and Mathjak's postings! Actually I agree with both of them and about 6 months ago reached a different conclusion. I have been investing heavily into investment grade prefered stocks that yield 6-7%. I have no interest in bonds at these rates. Yes, I understand Preferreds are not bond substitutes, but will act like them in some degree. The ones I have purchased are old illiquid issues that have been around from 20-50 years. Their current yield is very similar to what they were pre 2008 and their yield spread is historically high between them and government issues which should provide some protection. In fact even as the government bonds have recently ticked back up in yield, the preferred prices of mine actually have increased.
Besides, if/when they drop, I will just buy more and recieve a better yield.


Sent from my iPad using Tapatalk

[seraphim]

It's always a guessing game.

In Oct 2013 we had about $250k come available earmarked for bonds. It was in SV accounts earning about 2% - 2.25% annually. We left the money there, since it was earning about the same as a ten year treasury, and the value was stable. I decided to wait until interest rates and yields rose.

Had I put it n VBTLX with the rest of my bonds, it would have returned about 5.89% in 2014.

I think the Fed will permit the interest rates to rise slowly for minimum impact on the market - so I'm rethinking my strategy. Perhaps transfer $150k into VBTLX, or a similar active fund. I don't plan on touching bond funds for at least five years - maybe ten, depending - since I'd still have sufficient cash to carry me over for five years, if necessary.

Decisions, decisions....

I welcome any thoughts on the matter...

I'm retired with a 60s/40b target AA. I consider the SV accounts as part of the bond side, in planning. My WD is about 2%.


Sent from my iPad using Early Retirement Forum

[mathjak107]

The feds actions on the short end by raising the fed funds rate or discount rate have less than a coin toss chance of effecting investors bidding on bonds.

Where and how fast rates rise on bonds will be decided by the worlds investors not the fed's actions on the rates they control.