Your thoughts on Vanguard GNMA fund

Have a question about your personal investments? No matter how simple or complex, you can ask it here.
Post Reply
Topic Author
ShowMeMoney
Posts: 2
Joined: Mon Apr 11, 2011 4:55 pm

Your thoughts on Vanguard GNMA fund

Post by ShowMeMoney » Mon Apr 11, 2011 5:14 pm

Hello,

I'm contemplating whether I should buy vanguard GNMA fund in my taxable account. Is this a good idea? My criteria is that this investment should be pretty safe yet offer relatively high return, and I might need the money < 5 years.

Thanks in advance.

tibbitts
Posts: 9754
Joined: Tue Feb 27, 2007 6:50 pm

Post by tibbitts » Mon Apr 11, 2011 5:24 pm

I don't think there are any safe alternatives offering what most people would consider "relatively high return" for money you'll need in less than 5 years.

Paul

User avatar
nisiprius
Advisory Board
Posts: 40288
Joined: Thu Jul 26, 2007 9:33 am
Location: The terrestrial, globular, planetary hunk of matter, flattened at the poles, is my abode.--O. Henry

Post by nisiprius » Mon Apr 11, 2011 5:53 pm

1) The usual advice is to try to locate bond funds in a tax-advantaged account.

2) I am going to compare and contrast four Vanguard bond funds: the Vanguard GNMA Fund (VFIIX) you say you are interested in, the Vanguard Total Bond Market Index fund (VBMFX), the Vanguard Intermediate-Term Bond Index Fund (VBIIX), and the Vanguard Intermediate-Term Treasury Fund (VFITX).

I am going to make an unsophisticated comparison.

These are all intermediate-term, investment grade bond funds.

Vanguard puts all of them in the same place on their 1-5 risk scale: 2. Their scale isn't a precision instrument, but it means that because they are intermediate term, their share value fluctuates somewhat. They may differ in average credit quality but these are all investment-grade bonds, any default risk is very small and doesn't make much difference. You might be unpleasantly surprised if you think that these funds are just a money market fund on steroids. Vanguard characterizes the #2 risk level in this way:
Vanguard wrote:Conservative to moderate funds—Risk level 2

Vanguard funds classified as conservative to moderate are subject to low-to-moderate fluctuations in share prices. In general, such funds may be appropriate for investors with medium-term investment horizons (four to ten years).
So on the face of it, none of these funds is unsuitable for money you might need in five years.

Now I'm going to do my shtick, which is growth charts. These give you a visual indication of what you'd have seen on your brokerage statement if you'd put $10,000 into each of these four funds ten years ago. Please don't focus on "which won." Focus on their general behavior. And notice that the fund which came out ahead, VBIIX, also had the largest fluctuations and the biggest dip in 2008-9, the old "risk-reward" relationship.

Image

They all did roughly the same thing. They all showed reasonably steady growth, without the crazy ups and downs of a stock fund. They all grew about $10,000 to about $17,000 in ten years, something between 5% and 6% annualized.

For all of them, you almost always had more money at the end of each year than you did at the beginning of the year.

If you need this sort of fund, none of them would be a crazy choice.

But if you might need it in less than five, years, well, look at the size of those wiggles. If you can't hold on for the "4-7 years" Vanguard talks about, you might invest in one of these to get a higher return than you'd get in a bank CD, only to find that at the moment you needed the money you were in one of these downward dips. It's not like you'd lose half your money or anything like that. but one of those dips could easily wipe out a year's gains and you might well end up doing somewhat worse than if you'd just put it into a CD.

Which one is best, of course, is the subject of endless debate. The sort of Boglehead default choice--the one I use personally, just because I don't know any better--is the Vanguard Total Bond Market Index Fund, which in fact is, I dunno, something like 1/3 mortgage-backed securities. It's some of everything, it's one of Vanguard's nine suggested "core funds," it's the fund they use for their bond component in their Target Retirement Funds and so forth. In my personal opinion, it's the one to use unless you can articulate some specific reason--some reason other than past performance--for choosing something else.

Larry Swedroe, who is a genuine bond expert, does not like mortgage-backed securities. He recommends against Vanguard Total Bond Market Index Fund because it contains them. It's clear, though, that he's not talking crash or meltdown or anything like that. He just thinks its a suboptimal choice because of the prepayment risk and a technical issue called "convexity." He thinks there's a small amount of hidden risk, and that when the risk shows up people who think they're doing a tad better with their mortgage-based funds will discover they're doing a tad worse. If he doesn't like Total Bond, I assume he would dislike the GNMA fund.

That leaves you with the Intermediate-Term Index fund, which is roughly like VBMFX without the mortgage-backed components, and the Intermediate-Term Treasury fund, which is liked by people who like Treasuries and disliked by people who don't.

Between these four, I don't think the choice is very important. They are all going be much lower risk and much steadier, if slower growth than a stock fund. And on the face of it, Vanguard says they are appropriate for holding period of 4-7 years.
Last edited by nisiprius on Mon Apr 11, 2011 9:18 pm, edited 2 times in total.
Annual income twenty pounds, annual expenditure nineteen nineteen and six, result happiness; Annual income twenty pounds, annual expenditure twenty pounds ought and six, result misery.

User avatar
grabiner
Advisory Board
Posts: 26123
Joined: Tue Feb 20, 2007 11:58 pm
Location: Columbia, MD

Re: Your thoughts on Vanguard GNMA fund

Post by grabiner » Mon Apr 11, 2011 8:56 pm

ShowMeMoney wrote:I'm contemplating whether I should buy vanguard GNMA fund in my taxable account. Is this a good idea? My criteria is that this investment should be pretty safe yet offer relatively high return, and I might need the money < 5 years.
Unless you are in the 15% tax bracket, if you are holding the fund in your taxable account, you probably want Limited-Term or Intermediate-Term Tax-Exempt; this is likely to give better after-tax returns at about the same level of risk. (Note that GNMAs are taxable in most states, so the GNMA fund won't save you on state taxes as a Treasury fund would.)
Wiki David Grabiner

tibbitts
Posts: 9754
Joined: Tue Feb 27, 2007 6:50 pm

Re: Your thoughts on Vanguard GNMA fund

Post by tibbitts » Mon Apr 11, 2011 9:08 pm

ShowMeMoney wrote:Hello,

I'm contemplating whether I should buy vanguard GNMA fund in my taxable account. Is this a good idea? My criteria is that this investment should be pretty safe yet offer relatively high return, and I might need the money < 5 years.

Thanks in advance.
An important consideration in looking at historical returns is that there aren't a lot of examples of results starting from the interest rates we're at now. So while you aren't likely to experience a disaster by holding GNMA fund, whether you're going to experience what most of us would consider a favorable return is less clear. Not that there are any clear alternatives at the moment.

Paul

User avatar
abuss368
Posts: 18790
Joined: Mon Aug 03, 2009 2:33 pm
Location: Where the water is warm, the drinks are cold, and I don't know the names of the players!
Contact:

Post by abuss368 » Tue Apr 12, 2011 6:36 am

Depending on your tax rate, consider a municipal bond fund.

User avatar
Taylor Larimore
Advisory Board
Posts: 29187
Joined: Tue Feb 27, 2007 8:09 pm
Location: Miami FL

Excellent analysis

Post by Taylor Larimore » Tue Apr 12, 2011 6:56 am

Hi Nisiprius:

Thank you for your excellent and understandable analysis of various type bond funds.
"Simplicity is the master key to financial success." -- Jack Bogle

sschullo
Posts: 2545
Joined: Sun Apr 01, 2007 8:25 am
Location: Rancho Mirage, CA
Contact:

Re: Excellent analysis

Post by sschullo » Tue Apr 12, 2011 7:10 am

Taylor Larimore wrote:Hi Nisiprius:

Thank you for your excellent and understandable analysis of various type bond funds.
Hi Nisiprius,
Just wanted to add on to what Taylor sezs. I own all of those funds you analyzed in my tax deferred account knowing the duration consequences. My thinking exactly. Thanks
Public School K-12 Educators: "Ask NOT what your annuity sales person can do for you, ask what you can do to be a Do-It-Yourselfer (DIY)."

Sidney
Posts: 6746
Joined: Thu Mar 08, 2007 6:06 pm

Post by Sidney » Tue Apr 12, 2011 7:18 am

Larry Swedroe, who is a genuine bond expert, does not like mortgage-backed securities. He recommends against Vanguard Total Bond Market Index Fund because it contains them. It's clear, though, that he's not talking crash or meltdown or anything like that. He just thinks its a suboptimal choice because of the prepayment risk and a technical issue called "convexity." He thinks there's a small amount of hidden risk, and that when the risk shows up people who think they're doing a tad better with their mortgage-based funds will discover they're doing a tad worse. If he doesn't like Total Bond, I assume he would dislike the GNMA fund.
If I am not mistaken Larry has also pointed out that there are a lot worse "mistakes" people make -- like having bonds in the wrong place, buying individual stocks, investing in hedge funds, not having a plan .......
I always wanted to be a procrastinator.

Valuethinker
Posts: 39608
Joined: Fri May 11, 2007 11:07 am

Post by Valuethinker » Tue Apr 12, 2011 7:50 am

nisiprius wrote:
Larry Swedroe, who is a genuine bond expert, does not like mortgage-backed securities. He recommends against Vanguard Total Bond Market Index Fund because it contains them. It's clear, though, that he's not talking crash or meltdown or anything like that. He just thinks its a suboptimal choice because of the prepayment risk and a technical issue called "convexity." He thinks there's a small amount of hidden risk, and that when the risk shows up people who think they're doing a tad better with their mortgage-based funds will discover they're doing a tad worse. If he doesn't like Total Bond, I assume he would dislike the GNMA fund.

T.
Fantastic summary.

One quibble 'negative convexity'.

All bonds have convexity (the second derivative of price with respect to Yield to Maturity) ie the rate of change of the sensitivity of a bond (modified duration) with respect to a change in the Yield of the bond.

What's strange about mortgage backed securities (in the US) is that because mortgage borrowers can refibabce, when interest rates fall, bondholders get their money back sooner than they would like-- bad for bond investors and for the price.

On the other side you have extension risk. Interest rates go up, MBS bonds fall more than you would like, because mortgage borrowers do not refinance at the historic rates.

Negative convexity just means the bond price doesn't do what it is supposed to do (ie increase when interest rates fall; decrease when interest rates rise) or do it worse than for a conventional bond of the same coupon, term to maturity and credit risk (like a US Treasury Bond in this case).

livesoft
Posts: 70262
Joined: Thu Mar 01, 2007 8:00 pm

Post by livesoft » Tue Apr 12, 2011 8:02 am

Valuethinker wrote:Negative convexity just means the bond price doesn't do what it is supposed to do (ie increase when interest rates fall; decrease when interest rates rise) or do it worse than for a conventional bond of the same coupon, term to maturity and credit risk (like a US Treasury Bond in this case).
I agree that the GNMA fund does not do what it is supposed to do. Here is a chart of some of the favorite bond funds around here since October 2010. The GNMA fund is acting more like a short-term bond fund than an intermediate-term bond fund. It has lower volatility and better return than the favorites during the rising interest rates of the past few months. How do the GNMA detractors reconcile the actual facts with their statements?

Image

Full disclosure: I own the Vanguard GNMA fund, but I would not own it in a a taxable account.
Wiki This signature message sponsored by sscritic: Learn to fish.

Valuethinker
Posts: 39608
Joined: Fri May 11, 2007 11:07 am

Post by Valuethinker » Tue Apr 12, 2011 8:15 am

livesoft wrote:
Valuethinker wrote:Negative convexity just means the bond price doesn't do what it is supposed to do (ie increase when interest rates fall; decrease when interest rates rise) or do it worse than for a conventional bond of the same coupon, term to maturity and credit risk (like a US Treasury Bond in this case).
I agree that the GNMA fund does not do what it is supposed to do. .
What I have described is a feature of all MBS pricing. The open question is whether the additional yield compensates the investor for those risks (and the answer is that the option volatility, which drives that pricing, varies over time- -the spread you are paid to take on that risk varies over time).

Refi risk is dependent on non-interest rate factors as well, for example the refi rate on US MBS picked up significantly when it became possible to imply on line.

We haven't had a serious test of extension risk since the 1980s, AFAIK.

Valuethinker
Posts: 39608
Joined: Fri May 11, 2007 11:07 am

Post by Valuethinker » Tue Apr 12, 2011 8:19 am

livesoft wrote: I agree that the GNMA fund does not do what it is supposed to do. Here is a chart of some of the favorite bond funds around here since October 2010.
Curious why you would choose such a short time period to make your point?

GNMA fund data goes back to 1998? I don't think we have 1994 data for it, alas (should have done relatively well in 1994-- a black year for bonds, but cushioned by higher coupon rates and extension risk wouldn't really show up in a 1 year interest rate move).

Valuethinker
Posts: 39608
Joined: Fri May 11, 2007 11:07 am

Post by Valuethinker » Tue Apr 12, 2011 8:24 am

Sidney wrote:
Larry Swedroe, who is a genuine bond expert, does not like mortgage-backed securities. He recommends against Vanguard Total Bond Market Index Fund because it contains them. It's clear, though, that he's not talking crash or meltdown or anything like that. He just thinks its a suboptimal choice because of the prepayment risk and a technical issue called "convexity." He thinks there's a small amount of hidden risk, and that when the risk shows up people who think they're doing a tad better with their mortgage-based funds will discover they're doing a tad worse. If he doesn't like Total Bond, I assume he would dislike the GNMA fund.
If I am not mistaken Larry has also pointed out that there are a lot worse "mistakes" people make -- like having bonds in the wrong place, buying individual stocks, investing in hedge funds, not having a plan .......
In the hierarchy of 'mistakes' it's not a very serious one. I don't get too fussed about it in the context of VG TBM (25% of its bonds are MBS).

The bad thing with bonds (besides interest rate risk) is credit risk, and a bond issued by a US government agency with an explicit US Treasury guarantee has, by definition, no credit risk.

It could however catch the unwary at the wrong moment. Richard Bookstaber wrote a book 'A Demon of our Own Design' (he was head of risk at Morgan Stanley) and in a footnote (which I think you can access on google books) he specifically mentions the investment banks are asleep at the switch on risk from Mortgage Backed Securities.

That was a bullseye comment to me (and it came pre credit crunch). that the option premium (the amount you get paid to buy an MBS over a US Treasury bond) varies over time and that could lead to adverse price swings when you least want them.

huntertheory
Posts: 292
Joined: Sat Mar 27, 2010 8:05 am

Post by huntertheory » Tue Apr 12, 2011 8:27 am

Nisiprius: Excellent analysis. Is anyone else surprised though that the Intermediate treasury fund had essentially the same returns as the other three funds -- and in fact beat two of them over the ten-year period? I would have thought the risk/return relationship -- i.e. specifically the credit risk -- would have told me that the Treasury fund should have done the worst of the other three. Maybe the other three are invested in equally high grade securities, but I had thought the Treasury fund was all super triple-A (treasurys) whereas the others hover somewhere around AA or at least range from A to AAA.

Maybe I'm missing something (wouldn't be the first time!).

Valuethinker
Posts: 39608
Joined: Fri May 11, 2007 11:07 am

Post by Valuethinker » Tue Apr 12, 2011 9:23 am

huntertheory wrote:Nisiprius: Excellent analysis. Is anyone else surprised though that the Intermediate treasury fund had essentially the same returns as the other three funds -- and in fact beat two of them over the ten-year period? I would have thought the risk/return relationship -- i.e. specifically the credit risk -- would have told me that the Treasury fund should have done the worst of the other three. Maybe the other three are invested in equally high grade securities, but I had thought the Treasury fund was all super triple-A (treasurys) whereas the others hover somewhere around AA or at least range from A to AAA.

Maybe I'm missing something (wouldn't be the first time!).
It's a puzzle. It may just be our sample period doesn't show the risk/ return of the other securities.

I would have thought a GNMA fund would, over the long run, pay 0.5%-1.0% pa higher returns than a comparable US Treasury fund. It has not.

Reasons why (hypotheses):

- risk is less than I think, therefore market does not compensate

- last 20 years of falling interest rates is unique, and that, plus the credit crunch (tilting returns towards safe US government backed securities) means US MBS have done unusually well

- differences in average maturity ie yield curve effect (longer maturity = higher returns) between the 2 funds

- market is efficient in that it gives returns from 2 funds of identical credit risk, an identical return

I am not sure what the theoretically minded would think. These are 4 hypotheses I came up with.

User avatar
Opponent Process
Posts: 5157
Joined: Tue Sep 18, 2007 9:19 pm

Post by Opponent Process » Tue Apr 12, 2011 9:34 am

huntertheory wrote:Maybe I'm missing something (wouldn't be the first time!).
the thing that most people are missing is statistics. bottom line there are no significant differences between the funds in Nisi's graph. Nisi's point is to counter those that might tell you one of these is better or worse than the others. the simple fact is that they are virtually identical, though much discussion (theory) is generated trying to parse them apart. I love hypothesizing as much as anyone, I do it for a living, but I know when to give up looking for a significant difference and move on.
30/30/20/20 | US/International/Bonds/TIPS | Average Age=37

livesoft
Posts: 70262
Joined: Thu Mar 01, 2007 8:00 pm

Post by livesoft » Tue Apr 12, 2011 9:42 am

Valuethinker wrote:
livesoft wrote: I agree that the GNMA fund does not do what it is supposed to do. Here is a chart of some of the favorite bond funds around here since October 2010.
Curious why you would choose such a short time period to make your point?
A few reasons:
1. Folks are buying funds now, not in the 90s.
2. I want to predict the near future, not the long-term future.
3. Folks are worried about bonds (especially intermediate- and long-term) since interests rates are going up. China and the ECB have raised rates. US mortgage interest rates are going up.
4. The GNMA has a shorter duration than it has typically had.
5. The OP asked about a 5-year term.
6. The GNMA fund bested other intermediate-term funds in the time frame I chose.
7. One can always exchange out of GNMA at any point in the future.
Wiki This signature message sponsored by sscritic: Learn to fish.

beardsworth
Posts: 2135
Joined: Fri Jun 15, 2007 4:02 pm

Post by beardsworth » Tue Apr 12, 2011 10:11 am

I've read a lot of threads about GNMA Fund, and it seems that a lot of the points of view tend toward a "glass half–empty/glass half–full" kind of debate.

Those who say the glass is half empty will point to GNMA and say: "Look at it. When interest rates go up, people sit tight on their older mortgages and GNMA rates don't rise with the market as fast; and when interest rates go down, GNMA share prices don't appreciate as much as those of other intermediate–term bond funds."

Those who say the glass is half full respond: "Which is exactly why I like it––because, on average although not at every single moment of every year, it combines a generous income payout with less share price fluctuation than other intermediate–term bond funds."

Except for funds explicitly made up of direct Treasury–issued bills/notes/bonds (including TIPS), it's also the only non–Treasury vehicle known to me which nevertheless provides that Treasury–level "full faith and credit" guarantee (for those to whom it's important) for the underlying bonds in the fund.

I don't own it, but have been looking at it for a long time.

Marc

* * *
Edited once just to correct a typo.
Last edited by beardsworth on Tue Apr 12, 2011 2:39 pm, edited 1 time in total.

Valuethinker
Posts: 39608
Joined: Fri May 11, 2007 11:07 am

Post by Valuethinker » Tue Apr 12, 2011 10:53 am

MarcMyWord wrote:I've read a lot of threads about GNMA Fund, and it seems that a lot of the points of view tend toward a "glass half–empty/glass half–full" kind of debate.

Those who say the glass is half empty will point to GNMA and say: "Look at it. When interest rates go up, people sit site on their older mortgages and GNMA rates don't rise with the market as fast; and when interest rates go down, GNMA share prices don't appreciate as much as those of other intermediate–term bond funds."
Not quite. MBS go the *wrong* way against interest rate changes. You could experience *greater* volatility in price.

The data I saw said the average GNMA bond is paid off in 8 years. That's quite remarkable given the mortgages therein are 30 year mortgages. In my parents' day, that would have been unthinkable (refinancing didn't really exist). It gives you a very nasty duration problem on a large interest rate move.

Note large rate moves have a nonlinear impact over small ones-- the driving force is the implied volatility of that put option (the one you have written to the mortgage borrower, as an investor in the bond).

Those who say the glass is half full respond: "Which is exactly why I like it––because, on average although not at every single moment of every year, it combines a generous income payout with less share price fluctuation than other intermediate–term bond funds."
In bond speak, you are compensated for the higher risk by a higher Option Adjusted Spread (ie higher yield). But you are taking on higher risk-- that is how MBS are priced in the marketplace (and therefore the fund, made up of MBS).
Except for funds explicitly made up of direct Treasury–issued bills/notes/bonds (including TIPS), it's also the only non–Treasury vehicle known to me which nevertheless provides that Treasury–level "full faith and credit" guarantee (for those to whom it's important) for the underlying bonds in the fund.

I don't own it, but have been looking at it for a long time.

Marc
That is an argument for Fannie and Freddie bonds. Since they don't have the expicit guarantee, but are 'backstopped' by the US Treasury (which has nationalized them) there is a 'free lunch' in that they pay slightly higher interest rates (a matter of a few basis points now, I believe) fro what is, in effect, the same guarantee.

Perhaps the market is simply pricing in legislative uncertainty as to what happens next.

But PIMCO in particular made a *lot* of money on that trade-- bet big on Freddie and Fannie MBS, and indeed the issuers were taken under US government protection.

Default User BR
Posts: 7501
Joined: Mon Dec 17, 2007 7:32 pm

Post by Default User BR » Tue Apr 12, 2011 12:25 pm

Why is having a different response to changes in interest rates a bad thing? That would seem to present a diversification.



Brian

Valuethinker
Posts: 39608
Joined: Fri May 11, 2007 11:07 am

Post by Valuethinker » Wed Apr 13, 2011 3:53 am

Default User BR wrote:Why is having a different response to changes in interest rates a bad thing? That would seem to present a diversification.



Brian
Not if it is *worse* in both cases (rise and fall).

That's de-worsification.

The issue is whether the higher yield at purchase fully compensates the investor for that risk.

The historic data we have says GNMAs return about what IT US Treasury bonds return over the long run-- no risk premium.

Therefore either the risk doesn't exist (but GNMA bonds at issue are priced with estimates of that risk, and complex quantitative models to estimate it) or the market has not historically paid a return for it.

Larry S's point is that the risk will show up at the worst possible moment. Interest rates rise sharply, duration of the GNMA fund will move out against you. His chapter goes through the complexities in a nice and easy way.

If you want safety from bonds, GNMA bonds have little or no presumed credit risk (good in a repeat of 2008) and negative convexity (bad if a sharp change in interest rates).

keniles
Posts: 138
Joined: Sun May 18, 2008 3:34 am

evaluation

Post by keniles » Wed Apr 13, 2011 5:42 am

Nisiprius-wow! What a great summary you did. Thanks--Ken

Senin
Posts: 100
Joined: Sat May 14, 2011 3:42 am

Post by Senin » Mon May 16, 2011 3:22 am

Here is my take on GNMA's.

I was looking for a bond fund to compliment my stock mutual funds (80 stock/ 20 bond). So that my stocks can fly during the bull. But when the bear hits, I want to dollar cost out of the bonds into the stocks-- at a reduced price rate. But I need a bond that doesn't react badly to the bull market. What is that? GNMA.

Look at the chart from nisiprius. During the 08-09 disaster, the only thing going up is GNMA.

I like GNMAs.

Post Reply