Howdy, Stranger!

It looks like you're new here. If you want to get involved, click one of these buttons!

In this Discussion

Here's a statement of the obvious: The opinions expressed here are those of the participants, not those of the Mutual Fund Observer. We cannot vouch for the accuracy or appropriateness of any of it, though we do encourage civility and good humor.

    Support MFO

  • Donate through PayPal

Bond duration question

edited April 2021 in Off-Topic
Is it there? If yes, where? If not, could it be added when convenient?
«1

Comments

  • @WABAC.

    Thank you!

    Yes, in MultiSearch, you can specify Effective Maturity, which Refinitiv defines as:

    "Effective maturity is the measure of maturity of a portfolio’s bonds, taking into consideration the bonds may have call or put provisions. Years to maturity is calculated from next call date and data date for callable securities and from maturity date and data date for noncallable securities. Use the weighted average when calculating the portfolio-level effective maturity. Only uses the fixed income securities in the portfolio."

    There is also a break-out of bond duration allocations in the results table.

    Both found in the Portfolio/Bond Info group.

    A screenshot is here.

    But, me personally, my head hurts every time I try to make sense of bond duration metrics, especially when comparing say Lipper with M* and with the fund manager's own breakout/definitions.
  • Effective maturity is different from effective duration.

    I have issues with the definition above of effective maturity, but it will suffice to demonstrate that effective maturity and effective duration are different.

    Consider a single, vanilla, non-callable bond with a maturity of 20 years. According to the definition given, its effective maturity is 20 years: there are no call or put provisions to consider. Yet, if the bond has coupons, its effective duration is less than 20 years.

    Effective maturity is a measure of maturity (how long a bond is expected to remain outstanding) after adjusting for embedded options. Effective duration is a measure of interest rate sensitivity after adjusting for embedded options.

    See M*'s Fixed-Income Survey Guidelines, where on pp. 2-3 it defines these two terms. Its definition of average effective maturity is:
    a weighted average of the maturities of the bonds in a portfolio, taking into account all expected mortgage prepayments, puts, and adjustable coupons. Maturity is defined as the length of time in years until a debt becomes due for payment.
    http://advisor.morningstar.com/Enterprise/VTC/FI_Survey_Guidelines_cashRevision_FINAL.PDF

    That's pretty close to Refinitiv's definition.

    FWIW, an issue I have with Refinitiv's definition is that it draws no distinction between premium and discount callable bonds. A premium bond is likely to be called. A discount bond is likely not.

    Investopedia: "Knowing the likelihood that a bond may be called is crucial to computing average effective maturity."

    There are other factors that play into these calculations as well. For example an issuer that is in poor financial shape is not as likely to call a bond even if it is trading at a premium than is a financially sound issuer.
  • @Charles, what you describe does not seem to address interest-rate sensitivity as a bond screen. Thank you to MSF for explaining duration.

    An example of what I would like to do is screen for core bond funds with a portfolio rated => A, and a duration => 4 years and <= 5 years. Then I know the managers -- or indexes -- that show up in the results are operating on roughly similar terrain when I look at performance, Sharpe, secret sauce, etc.

    Thank you for getting back to me.

  • edited April 2021
    Duration is a lot more accurate with treasuries. Many/Most bond funds have several bond categories that behave differently, some use derivatives which complicate things even more. Managed funds can also change their bond mix which affect performance.
    I can show plenty of funds with shorter duration that lost more on rate changes and why duration isn't a top criterion when I look at managed funds.
    Basically, it would be harder than you think to predict future performance based on duration and quality unless you select treasury funds.
  • Duration works fine so long as you understand what it represents.

    The price of a bond moves contrary to the movement of the interest rate for that bond. If you're trying to figure out the price change of a 1 year T-bill by looking at changing rates in 10 year Treasury notes you'll likely be far off. Interest rates even for treasuries don't all shift by the same amount.

    Likewise, if you're interested in the price change of a BB 10 year, look at the junk bond yield curve movement, not how treasury rates are doing. Interest rate curves for different types of bonds don't shift in parallel.

    Next, one also needs to recognize that duration is just the slope of the tangent to the appropriate price/yield curve. Emphasis on curve. Duration is just a linear approximation. The more quickly the curve bends, the faster the price deviates from the tangent. That's why it is important to look not just at the first derivative of the curve (duration) but the second derivative (convexity).

    image

    This is especially true of bonds with negative convexity, notably MBSs. Their convexity is often negative. When convexity is positive, as rates go higher, bond prices drop, but they drop less than they would following a straight line. When convexity is negative, prices drop faster.

    No rocket science here. Effective duration, along with looking at the appropriate interest rate curve and being mindful of convexity can give an investor a good sense of what to expect from a bond portfolio.
  • edited April 2021
    3 examples:
    Let me know how you can predict PIMIX movements based on its duration.

    DODIX is simpler, but the managers can still change within category. Imagine they reduce treasuries % and raised Corp.

    Even a simple fund such as VBTLX wouldn't be an easy/trivial task because it has 3 main bond categories.
  • msf
    edited April 2021
    Duration is a lot more accurate with treasuries.

    This was the claim. Two actually:
    1. Duration is accurate with Treasuries
    2. In comparison, duration is inaccurate with any other type of bond

    Nothing here about mixing different types of bonds, nor did I say anything about that. In fact, I wrote about individual bonds: "the price change of a 1 year T-bill", and "price change of a BB 10 year". So let's do away with the distractions and get down to fundamentals.

    1. Even with Treasuries, duration alone isn't going to tell you which Treasury will lose more money when "interest rates" rise 1%. (I put interest rates in quotes because points the Treasury yield curve - yield vs. maturity - do not rise or fall in unison. The curve may steepen or become more shallow. But for the sake of argument, let's assume the rates for all Treasuries rise 1%)

    Here are two real Treasury bonds, CUSIP 912810SD1 (duration 19.157 years) and CUSIP 912810RT7 (duration 19.118). While the durations are close, the former clearly has a longer duration than the latter. One would expect the former to drop more in price should "interest rates" rise 1%. One would be mistaken.

    I ran these real bonds through a bond price calculator. The former currently yields 2.326% based on its open market price. If the market rate for this bond rises to 3.326%, its price drops 17.10%. With this alone we can see that duration isn't all that accurate, it price drop predicted was 2% too high.

    But it gets worse. The latter currently currently yields 2.34%. If the market rate for this bond rises to 3.34%, the price of this (slightly) shorter duration bond drops 17.13%, a tad more than the longer duration bond.

    The reason is actually pretty clear. The former has a (positive) convexity of 463, while the latter has a convexity of "only" 443. The greater the magnitude of convexity the greater the deviation between the straight line duration-predicted price change and actual price change.

    You can run the calculations yourself.

    2. The entire exercise above said nothing about Treasuries. The only inputs to the price calculator are: coupon, current market rate (or current market rate + 1%), and maturity. (This calculator also uses days since last payout, but that's to compute the dirty price and we're talking about clean prices here.) Well, I did also have to say that coupon payments were semi-annual.

    Any type of vanilla bond, even a vanilla junk bond, would produce the same results. So the (in)accuracy of duration in predicting price movement is the same regardless of the type of bond.

    Of course calculating duration does get more complicated once one moves away from vanilla bonds. Embed puts as with MBSs, or make a bond callable, and the calculation of effective duration has to incorporate the possibilities (and probabilities) of what might happen. After all that, all one has for duration is just an expectation value.

    BTW, you'll notice that I selected two Treasuries with coupons. I didn't select a zero because technically that's a derivative, albeit one where the calculations become simpler, not more complex. Had I used a zero, the numeric comparisons would have been more stark.
  • edited April 2021
    msf, nice try. I never said :In comparison, duration is inaccurate with any other type of bond".

    I sais "Many/Most bond funds have several bond categories that behave differently, some use derivatives which complicate things even more. Managed funds can also change their bond mix which affect performance."
    I posted about bond funds, not a vanilla single bond

    Please relate to the entirety of my comments and contexts and the examples I gave PIMIX, DODIX, VBTLX. You can't predict in advance PIMIX, DODIX and even VBTLX wouldn't be an easy/trivial task.

    Start over

    Lastly, I made tons of money using bond OEFs paying little attention to duration.
  • msf
    edited April 2021
    I never said :In comparison, duration is inaccurate with any other type of bond".

    You said that "Duration is a lot more accurate with treasuries". Thus, duration is a lot less accurate with non-treasuries.

    Then you went on to write about managed funds, which has nothing to do with treasury funds vs. non-treasury funds. For example, VFITX is a managed treasury fund.

    VFITX can hold up to 15% in illiquid securities. That is, it "can change [its] bond mix which affect[s] performance. It also holds "derivatives which complicate things even more." Less than 90% of its portfolio is government debt, let alone Treasuries. Not all of that government debt is even US government debt.

    As you wrote, to "predict in advance PIMIX, DODIX, and even VBTLX wouldn't be an easy/trivial task." That much is true - the more complex the mix, the harder the task or sorting out the pieces. Even for a pure 100% treasury fund (or for that matter a pure, 100% IG corporate bond fund) it wouldn't be a trivial task:

    You can't just multiply "the" interest rate change by the duration because rates change differently for bonds of differing maturities. Currently the yield curve is steepening - interest rates are rising more slowly for short term bonds than for long term bonds. Since two portfolios can have the same duration with different mixes of longer and shorter maturity bonds, you can't just multiple the average duration by some mythical "average" change of interest rates. You have to weight the interest rate changes by the distribution of bonds in the fund's portfolio. Not a trivial task.

    Add to that the fact that any information you have about a fund's portfolio is at least 30 days out of date.

    In short, you're conflating fund "purity" with fund type. There's little special about treasuries, and nothing that you've identified.

    Start over.
  • edited April 2021
    msf: "In short, you're conflating fund "purity" with fund type. There's little special about treasuries, and nothing that you've identified."

    wow, I'm "impressed".
    I have done the following a few times. I looked at treasuries rate changes when it was rapid and used VG bond funds and found out that treasury funds price changes were pretty accurate compared to others. I and probably most investors are looking at a big picture view.

    Maybe you can answer the following why Muni can be completely different. YTD: rates are up which means most simple bond categories are down, but Munis are up. I know Munis have tax advantage, but are they up YTD? If you look at a typical Munis (here) it does NOT explain it.

    Me? I do the usual, select several funds/categories based on a nice smooth uptrend chart and why I used Muni fund since 04/2020 and prior to 03/2020. I hardly ever use duration.
  • You're making it too easy.

    Repeating what I wrote before: "Likewise, if you're interested in the price change of a [muni] 10 year, look at the [muni] bond yield curve movement, not how treasury rates are doing. Interest rate curves for different types of bonds don't shift in parallel."

    From R.W. Baird (dated March 8):
    Bottom Line:
    •Treasury yields rose some more hitting the highest level since February 2020.
    • Muni yields fell after a surge since mid-February. The Bloomberg Barclays Muni Index is now ~32 bps higher since the lows about a month ago.
    https://content.rwbaird.com/RWB/Content/PDF/Insights/Municipal-Bond-Market-Commentary.pdf

    Again repeating the obvious: "The price of a bond [fund] moves contrary to the movement of the interest rate for [those] bond[s]."
  • FD1000 said:

    ms

    Me? I do the usual, select several funds/categories based on a nice smooth uptrend chart and why I used Muni fund since 04/2020 and prior to 03/2020. I hardly ever use duration.

    I've been reading your bond posts for about ten years.

    Just about every bond performance chart you post includes a column showing duration.

    You routinely discuss duration as a critical dashboard metric for bond investing.

    Today you posted this on another site (bold added):

    https://armchairinvesting.freeforums.net/thread/616/bond-oefs-2021?page=16&scrollTo=11522

    I get questions about Munis, at this time I prefer to stay within shorter term duration NVHAX because it did very well YTD + I can't predict rates direction (likely going up).

    Your statement that "(You) hardly ever use duration" is simply not true.
  • edited April 2021
    Sure, stillers(=alb..) the usual, your posts are out of context.
    Do I invest based on duration? I sure do post lots of info with my monthly info but my style for years is still based mainly on momentum + smoother ride.

    But I also post generic answers for generic questions. So, yes, duration is more important for simpler bond funds but much less for complex funds which I posted about many times.

    BTW, how are your 4 funds to replace 3% CD doing so far YTD? (link)
    YTD...VSCSX -0.4%...IBFFX -1%...THIFX -1.2%...VBIRX -0.5%...LALDX +0.7%

    But wait, what happened to your forecast from 2017-18 that my investment style is bogus, I will not retire and will not make it? Reality: in the last 3 years I made 12.5% annually + SD=3.5. Try to find any fund with that risk/reward or Sharpe.
  • edited April 2021
    fd1k, again, you really should do this for a living for others and not bother poasting
  • edited April 2021

    fd1k, again, you really should do this for a living for others and not bother poasting

    Why would I bother working? After you have enough, why have a schedule, customers and bosses.
    My grandson says "grandpa is retired, he plays all day"...he nailed it:-)
  • If you want to learn how to debate, then read msf.
    If you want to learn how to trade, then read FD1000.

    msf can write 10,000 words and I can't make a penny from them.
    FD1000 can write 10 words and I can make a fortune.

  • FD1000 can write 10 words and I can make a fortune.

    Sounds like you 2 should get your own (chat) room.
  • If you want to learn how to debate, then read msf.
    If you want to learn how to trade, then read FD1000.

    msf can write 10,000 words and I can't make a penny from them.
    FD1000 can write 10 words and I can make a fortune.


    you mean HE can make (always does make) a fortune
  • I like data. That's why I asked about duration.

    I spent some of my work life selling data to people looking for answers. So I'm aware of the caveats.

    Pretty funny how this thread turned out.
  • Great thread! I’m learning who the giants of the investing universe are.
  • JD_co said:


    FD1000 can write 10 words and I can make a fortune.

    Sounds like you 2 should get your own (chat) room.
    Sounds like you can't face reality. If you are interested in making money while achieving great risk-adjusted performance pay attention.
  • edited April 2021
    FD1000 said:

    JD_co said:


    FD1000 can write 10 words and I can make a fortune.

    Sounds like you 2 should get your own (chat) room.
    Sounds like you can't face reality. If you are interested in making money while achieving great risk-adjusted performance pay attention.
    Nobody likes to hear some blowhard who likes to brag - hiding behind a keyboard, no less.
    Perhaps you could take your talents to M*.
  • Hey Team, let's not get side tracked by non issues..who cares how one presents the results of their investing prowess....keep the focus on trying to educate, learn from others, share viewpoints, life and investing experiences and offer support to each other in a respectful fashion and maybe have a few laughs along the way.

    I stopped looking at the other board at M*..too much of snippy comments back and forth with strangers on the internet...really a time waster...who knows we could live down the block from each other or 1500 miles away, the majority of us don't have a clue who each other are so...really...who gives a sheet?

    On a side note...just increased my holdings in Home Depot to low six figures...

    They are going to benefit with the raw material increases, multi family developments going up all across the country, too expensive to build and buy single family homes...all going to drive their top line biggly...and me thinks they will be selling a lot of building materials soon....you can feel the riots coming on especially with warmer weather and the trial in MN along with the recent tragic incident...why try to disrepect the cops and run away cause you have something on you...why shoot another human being cause he has something on him...as far as I know he didn't hurt anyone else, rob anyone else...so why shoot or Taser the guy, he didn't strike any cops, was just squirming and worming to get away...even if it is accidently? Kind of familiar to the guy in Kenosha...disrepecting the cops, not following lawful orders and thinking he can just do what he wants and leave, then blamo. What a mess.

    Best to all,

    Baseball Fan
  • edited April 2021

    On a side note...just increased my holdings in Home Depot to low six figures.

    Looks like you moved the stock. Home Depot gained a buck and a half today. (HD) 320.79 USD +1.56 (0.49%)
  • JD_co said:

    FD1000 said:

    JD_co said:


    FD1000 can write 10 words and I can make a fortune.

    Sounds like you 2 should get your own (chat) room.
    Sounds like you can't face reality. If you are interested in making money while achieving great risk-adjusted performance pay attention.
    Nobody likes to hear some blowhard who likes to brag - hiding behind a keyboard, no less.
    Perhaps you could take your talents to M*.
    done and done, I believe
  • edited April 2021
    JD_co said:

    FD1000 said:

    JD_co said:


    FD1000 can write 10 words and I can make a fortune.

    Sounds like you 2 should get your own (chat) room.
    Sounds like you can't face reality. If you are interested in making money while achieving great risk-adjusted performance pay attention.
    Nobody likes to hear some blowhard who likes to brag - hiding behind a keyboard, no less.
    Perhaps you could take your talents to M*.
    I don't care what you like or not. I have been posting for many years about all type of mutual funds. In the last several years I mainly invested in bond OEFs because it works extermly good for me. I have similar threads like the one on MFO (link) in several sites (been on M* over 10 years).
    Maybe you can learn something new instead of complaining. There is a good reason why good posters stop or hardly post anymore as I did with the above thread.
  • Aside from everything else going on in this thread, I would like to see any Lipper data on duration. And I would like to be able to screen on their duration numbers the way I can screen on many other factors.



  • edited April 2021
    WABAC said:

    Aside from everything else going on in this thread, I would like to see any Lipper data on duration. And I would like to be able to screen on their duration numbers the way I can screen on many other factors.

    +1

    I hope you get your data. However, if the thread continues to veer far off topic, you can easily shuffle it over to the “off-topic” section where playground bragging really belongs.

  • hank said:

    WABAC said:

    Aside from everything else going on in this thread, I would like to see any Lipper data on duration. And I would like to be able to screen on their duration numbers the way I can screen on many other factors.

    +1

    I hope you get your data. However, if the thread continues to veer far off topic, you can easily shuffle it over to the “off-topic” section where playground bragging really belongs.

    Excellent suggestion. Thank you.
  • edited April 2021
    Actually - This thread is better called “Former Bond Duration Question” - after devolving into a sniping match over who the best investor on the board is.:(
Sign In or Register to comment.