Actively managed vs index bond funds

[VasyaPupkin]

In discussions with a financial adviser of work's 401k, and while questioning rather steep fees, I've been told that the resulting performance more than covers that. As an example, the following Vanguard funds were given as a comparison to the expensive (1.74%)  fund being "pushed":

Vanguard Inflation-Protected Securities Fund  VIPSX   -8.83%
Vanguard Total Bond Market Index Fund  VBMFX -3.24%
Vanguard Intermediate-Term Bond Index Fund  VBIIX -4.70%
Vanguard Long Term Bond Index Fund VBLTX  -11.09%

Pioneer Strategic Income Fund  PSRCX -1.13%
Looking at the returns, it certainly appears that Pioneer fund is better. Digging deeper I see that it has higher percentage of cash compared to the rest. Is that all there is to this perceived difference? If long term we should be seeing better performance from index funds vs expensive active ones, what would explain the opposite seen in the graphs? Or am I missing something critical? My feeling is that graphs shown on morningstar are not what I as an individual investor will actually see by the time all fees kick in. Or this particular fund is just "being lucky" for a bit

[ncornilsen]

What time period are those returns based on?

[beltim]

There's long been a sense in the finance world that the really smart and capable people go into fixed income rather than equities.  Having never worked on Wall Street, I can't verify this personally.

I haven't seen similar studies on bond index funds as I have on stock index funds, and I don't know why.  But I can say that most of the indices that bond funds track are considerably quirkier.  For one thing, there is necessarily a lot more turnover in bond indices. You might look at the criteria for inclusion on those Vanguard funds and see how bonds are selected for inclusion.

All this is a long way of saying I don't know, although there are some good reasons for bond index funds to behave very differently than stock index funds. 

[arebelspy]

I'm guessing those are cherry picked results.  Where the heck do those numbers come from?

How long has the fund been around?  What is its long term track record, and what makes you think they can continue to outperform in the future?

[beltim]

Those are very close to trailing one year returns.  I'd bet those are trailing 1 year returns from when Vasya went to the financial planner. 

It looks like PSRCX has better 3, 5, and 10 year returns than the funds excepting  VBLTX, which has very closely tracked PSRCX.  It has much lower turnover than the Vanguard funds, too.

[arebelspy]

Edit: Duh, I misread.  My bad.

[fiveoclockshadow]

Trailing returns of an actively managed fund tend to be meaningless. Remember, over a period of many years some funds will outperform an index. The problem is that looking forward an actively managed fund that has performed well in the past is just as likely to perform poorly going forward.

Buffet long ago made an analogy to "coin managers". Imagine a thousand managers flipping coins. You expect a few to flip many heads in a row. Should you suddenly buy their coin fund?  Would you expect them to get more heads going forward than any other manager?

As to the myth that active management is better in fixed income the data is clear - no it isn't. Bogle has a whole chapter in Common Sense and the annually updated data shows active bond management still poor. The best estimator is still cost. So a low cost active bond fund (they exist) would probably be just fine compared to an index. But a high cost one is a losing proposition.

[HamhockHammock]

Bogle has a whole chapter in Common Sense and the annually updated data shows active bond management still poor. The best estimator is still cost. So a low cost active bond fund (they exist) would probably be just fine compared to an index. But a high cost one is a losing proposition.

Bogle also discusses it a bit this podcast. http://www.econtalk.org/archives/_featuring/john_bogle/. I also can't recommend Econtalk enough. I always learn something from it.

[aj_yooper]

Bogle has a whole chapter in Common Sense and the annually updated data shows active bond management still poor. The best estimator is still cost. So a low cost active bond fund (they exist) would probably be just fine compared to an index. But a high cost one is a losing proposition.

Bogle also discusses it a bit this podcast. http://www.econtalk.org/archives/_featuring/john_bogle/. I also can't recommend Econtalk enough. I always learn something from it.

Thank you for the link!  Excellent interview.

[VasyaPupkin]

I just finished reading the common sense book myself. For the long term results I've been looking at the hypothetical return curves for 10k invested x number of years using Morningstar web page

[kyleaaa]

Under absolutely no circumstances should you buy that fund. The Vanguard funds aren't even close to comparable, as they are all high-quality investment/government-grade bond funds.

PSRCX  is basically a junk bond fund that lost over 12% in 2008! That compares to a 5+% gain for the Vanguard funds in that same year. This fund is not safe.

[beltim]

Under absolutely no circumstances should you buy that fund. The Vanguard funds aren't even close to comparable, as they are all high-quality investment/government-grade bond funds.

PSRCX  is basically a junk bond fund that lost over 12% in 2008! That compares to a 5+% gain for the Vanguard funds in that same year. This fund is not safe.

That just means it invests in different kinds of bonds.  If the OP wants safer bonds, I agree with you, but if they want junk bonds, then the Vanguard funds listed don't make sense.

[beltim]

Under absolutely no circumstances should you buy that fund. The Vanguard funds aren't even close to comparable, as they are all high-quality investment/government-grade bond funds.

PSRCX  is basically a junk bond fund that lost over 12% in 2008! That compares to a 5+% gain for the Vanguard funds in that same year. This fund is not safe.

That just means it invests in different kinds of bonds.  If the OP wants safer bonds, I agree with you, but if they want junk bonds, then the Vanguard funds listed don't make sense.

But, of course, Vanguard offers a high yield fund as well: VWEHX.  It's outperformed all of the listed funds above over 1, 3, 5, and 10 year periods.  And it has an expense ratio of 0.23%, significantly lower than the Pioneer fund listed earlier, and likely the reason for its outperformance. 

[beltim]

I have a question for the general public here.  The Vanguard funds do way more buying an selling than one would expect in an index fund.  In a simplistic way of looking at it, the turnover should be roughly 1/(2 * average maturity).  So it the average maturity is 10 years, one would expect annual turnover of about 5%, as 1/20 of the bonds expire each year since about half the bonds have maturities less than 10 years and half have maturities greater than 10 years.  I can see various ways where this simplistic analysis would be off by a factor of 2 or so. 

However, the Vanguard "Index" funds have much higher turnovers.  VBIIX, the intermediate term fund, has an average maturity of 7.3 years but a turnover of 65%.  So their average bond has a maturity of 7.3 years, but they only hold it for 1.5 years.  Why is this?  I find it hard to believe that the index completely turns over every 18 months (I tried to find this data but couldn't - the relevant index is the Barclays U.S. 5–10 Year Government/Credit Float Adjusted Index for anyone who wants to look).

[kyleaaa]

I have a question for the general public here.  The Vanguard funds do way more buying an selling than one would expect in an index fund.  In a simplistic way of looking at it, the turnover should be roughly 1/(2 * average maturity).  So it the average maturity is 10 years, one would expect annual turnover of about 5%, as 1/20 of the bonds expire each year since about half the bonds have maturities less than 10 years and half have maturities greater than 10 years.  I can see various ways where this simplistic analysis would be off by a factor of 2 or so. 

However, the Vanguard "Index" funds have much higher turnovers.  VBIIX, the intermediate term fund, has an average maturity of 7.3 years but a turnover of 65%.  So their average bond has a maturity of 7.3 years, but they only hold it for 1.5 years.  Why is this?  I find it hard to believe that the index completely turns over every 18 months (I tried to find this data but couldn't - the relevant index is the Barclays U.S. 5–10 Year Government/Credit Float Adjusted Index for anyone who wants to look).

Your expectation is incorrect. It is not reasonable to assume about half of the bonds in the portfolio have maturities less than 10 years and half have maturities more than 10 years. Perhaps the fund holds a few 30 year bonds and a ton of 1-year bonds to balance them out. This is not uncommon.

Additionally, the turnover accounts for things like fund flows, ratings changes (an investment-grade index can't hold a bond if it drops below investment grade and vice versa), and index changes.

[beltim]

I have a question for the general public here.  The Vanguard funds do way more buying an selling than one would expect in an index fund.  In a simplistic way of looking at it, the turnover should be roughly 1/(2 * average maturity).  So it the average maturity is 10 years, one would expect annual turnover of about 5%, as 1/20 of the bonds expire each year since about half the bonds have maturities less than 10 years and half have maturities greater than 10 years.  I can see various ways where this simplistic analysis would be off by a factor of 2 or so. 

However, the Vanguard "Index" funds have much higher turnovers.  VBIIX, the intermediate term fund, has an average maturity of 7.3 years but a turnover of 65%.  So their average bond has a maturity of 7.3 years, but they only hold it for 1.5 years.  Why is this?  I find it hard to believe that the index completely turns over every 18 months (I tried to find this data but couldn't - the relevant index is the Barclays U.S. 5–10 Year Government/Credit Float Adjusted Index for anyone who wants to look).

Your expectation is incorrect. It is not reasonable to assume about half of the bonds in the portfolio have maturities less than 10 years and half have maturities more than 10 years. Perhaps the fund holds a few 30 year bonds and a ton of 1-year bonds to balance them out. This is not uncommon.

Additionally, the turnover accounts for things like fund flows, ratings changes (an investment-grade index can't hold a bond if it drops below investment grade and vice versa), and index changes.

That can only be true for index funds if the index has a similar distribution of maturities.  It is of course quite common in actively managed bond funds.

Even so, say I'm off by a factor of 2.  So normal bond turnover is 1/(1*maturity) or, rounded up, 14%.  Fund flows shouldn't affect the turnover unless there are withdrawals.  Index changes are hard to gauge since I can't find the actual index (I know many of these indices are created specifically so that funds can track them), but logic suggests that it should be similar to new issuance, so that's another 14%.  Ratings changes are an interesting point, but with our other estimates that would suggest that ratings changes would have to happen for >37% of the bonds in the fund, which stretch credulity.

The 65% for this fund isn't unusual, by the way: over the past 5 years, annual turnover has ranged from 46% to 89%.

[fiveoclockshadow]

You'll want to read this:

http://isharesblog.com/blog/2011/09/13/qa-on-bond-funds-and-churn-why-turnover-can-be-misleading/

Bond index fund turnover is nothing at all like equity index fund turnover.  The VBIIX turnover is typical of a bond index fund.  Nothing amiss.

[beltim]

You'll want to read this:

http://isharesblog.com/blog/2011/09/13/qa-on-bond-funds-and-churn-why-turnover-can-be-misleading/

Bond index fund turnover is nothing at all like equity index fund turnover.  The VBIIX turnover is typical of a bond index fund.  Nothing amiss.

Thanks for the link.  It's a bit basic, but it does have some good info: I did not know that bond indices rebalance monthly.  Is there any literature to justify the indices themselves changing so often? 

An actively managed fund might very easily be able to outperform an index simply by reducing fees thanks to fewer transaction costs.  As you said earlier, cost is the best predictor of performance, not active vs. passive.

[fiveoclockshadow]

Thanks for the link.  It's a bit basic, but it does have some good info: I did not know that bond indices rebalance monthly.  Is there any literature to justify the indices themselves changing so often? 

Well the index itself justifies them changing, right?  It is an intermediate bond fund, thus it must hold substantially intermediate bonds.  Note, that doesn't mean on average the maturity comes out intermediate - the individual bonds held must be intermediate term.  Thus the fund can not, by definition, hold the bonds until maturity.  That's why your simple equation doesn't work.  And the SEC mandates that if you put the word "intermediate" in the name of the fund you must hold mostly intermediate bonds - hence active vs. passive doesn't matter.  You will have a lot of turnover just on maturity alone.

Next for corporate bonds there are calls - the bonds don't even exist until maturity in many cases.  They are called and you must buy new ones with the returned capital.

Finally there are *vastly* more bonds than there are equities.  Constructing a index for equities is trivial compared to that for bonds, the composition changes more slowly for equities (there are a few IPOs per year and a few de-listings).  Bonds are issued and disappear all the time on the other hand.  The bond index updates monthly not arbitrarily but because the market is changing that fast and for the index to actually track it needs frequent updating.

In summary they are different because the bond market and equities markets are not at all similar in any way, shape or form.  Bonds are issued and called far, far, far more frequently than shares of equity are.