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.
Is there any reason to feel optimistic about Templeton's Global Bond Fund (TPINX) and Global Total Return Fund (TGTRX)? Their recent performance is very weak, but would it make sense to hold on?
A lot depends, I guess, on why you bought Hasenstab in the first place. He's a very independent investor, rare in the world of fixed-income investing. He tries to get ahead of the crowd, which paid off well in his bets on Irish and Hungarian bonds. So far his investment in Ukrainian bonds has been ugly. In term of short term performance, a larger issue has been his attempt to position his funds to deal with U.S. interest rate hikes. He still thinks we'll see them this fall. Ahead of that, he's moved a quarter or so of his funds in cash and purchases some derivatives which will rise if interest rates do.
If you were looking for meek 'n' mild, it was probably a misplaced bet to begin with. If you're looking for "always wins," all bets are misplaced.
In general, he badly trails his benchmark about one quarter in four then briskly leads it in the next two. That's been the case so far in 2015: down in 2014Q4, ahead in 2015Q1 and 2015Q2.
I was a long-term holder of TTRZX & TGBAX (it was ~9% of my portfolio at one time). I sold, shortly after I became aware of his Ukraine bets, which was preceded by some time of underperformance.
I came to ask myself: would I, on my own, invest in ANY Ukrainian bonds? (not a chance, not a dime) Do I really want or need to be taking risks with the bond portion of my portfolio, to the extent that he was? I'd rather take the risks on the equity side of my portfolio.
I think the nature of some of the risks he take, are philosophically, not the kind I am prepared to take with bonds. My sense is that over the years he has gotten more "swing for the fences" in making bets. I'm not too comfortable with that type of manager in equities, let alone in bonds. But that is just me.
I've held TGBAX for the past 6+ years. Pretty solid fund that at times acts like a stock fund. I generally like it but waffle about keeping it sometimes. As for the Ukraine impact, the fund's got like 2.5% exposure to the country, which to me is hardly worth worrying about, frankly.
IMHO there are two reasons for holding bonds - periodic interest/dividends, and diversification. These days, I don't see bonds doing much better than cash in the first (without taking on a lot more risk, which sort of defeats the second objective).
So it seems the main reason to hold a bond fund is as an equity diversifier. TGBAX doesn't serve the role quite as well as a vanilla bond fund in this regard, but it does it with (as rfono observed) equity-like performance. 3 year correlation coefficients (R^2 is, well, the square of these numbers) are 0.54 for TGBAX vs. VFINX, and 0.28 for VBMFX vs. VFINX.
There's nothing else quite like Hasenstab's funds, so I think they're great as diversifiers. That includes currency.
Hasenstab almost completely decouples interest rate positions from currency positions. He can be long foreign bonds (for better yields) while short foreign currency. That's part of the appeal of his management style. Right now, TGBAX is 78% long in US dollars (more than double the benchmark Citigroup World Gov. Bond Index), and short Euros (24%) and Yen (36%).
If Fuss, Ivascyn, Gundlach, Rivelle and gals...Gaffney are doing better than Hasenstab at this time why not take some or all of his fund money and move it into one of the better performers. You are losing the value of time compounding by this misplaced loyalty. prinx
What time frame are you thinking about when talking about performance "at this time"? YTD? 1 year, 1 month?
If some of these other managers are doing even worse at this time, would staying with them also be misplaced loyalty?
TGBAX is down 1.75% YTD. LSBDX (Fuss) is down 3.62%. His protege is doing even worse - down 6.04% at EVBIX. (Data from M*, as of Aug 7th.)
It seems to me that loyalty is not misplaced if long term managers have demonstrated superior performance and discipline over time, and circumstances have not changed. Are these managers doing worse than would be expected for their various fund categories?
If not, then we're not talking about disliking the manager, but disliking categories. So perhaps we should be discussing asset allocation, not manager performance.
If Fuss, Ivascyn, Gundlach, Rivelle and gals...Gaffney are doing better than Hasenstab at this time why not take some or all of his fund money and move it into one of the better performers. You are losing the value of time compounding by this misplaced loyalty. prinx
Because not everyone wants to chase performance and/or do so very frequently and/or at the first whiff of not meeting their benchmark? A 1.75YTD loss compared to others in the same category that are down 3, 4, 6 or more percent is not a concern to me, and suggests either a) luck or b) manager being careful.
Had invested monies with Hasenstab; but sold this holding 3 years ago. Sold all of LSBDX in January. Although LSBDX has had some equity inclusion. More to do with this investment area (global bonds) in general and not favoring one manager over another manager. Edit note: I incorrectly noted a 20% equity position in the past with LSBDX. @msf has corrected this with his below report. Thank you, msf.
"Although LSBDX has had about a 20% equity inclusion."
The fund has tended to add a little equity in recent years, but nothing over 10%. A higher equity position must have been a long time ago. Here's what I found in the (semi) annual reports over the past few years:
Actually, TGBAX is doing better than Fuss and Gaffney, not as well as Ivacyn and the Poobah. But why anyone would compare him with these is beyond me. Nothing has changed with the way TGBAX is managed. What HAS changed is the various world economies, and the dollar in relation to other currencies. Investors who bought this fund should have bought it for the diversification it adds to a portfolio, not for the returns that it had in any given year or time frame. Hastenstab is incredibly defensive right now, with a duration of 0.13 years and an average maturity of 2.36 years. We captured some significant gains a year or so ago, but we maintain our positions in the fund.
To be fair to the thread, I praised Hasenstab along with these other bond fund managers. My comparing them as "great" had more to with their ability to manage risk.
In the world of investing, investors often falls victim to the most common denominator of reward over shorter and shorter periods of time when what makes bond manager's great may have more to do with the way they manage risk over longer and longer periods of time.
I have considered pairing a fund like TGBAX with an uncorrelated world equity fund.
Pairing TGBAX with Vanguard's Global Minimum Volatility Fund (VTWSX) might be a conservative way to gain that "balanced bond/equity" exposure to world investments.
PYGFX has done a good job. It is truly global, with 40-50% in U.S. bonds. TGBAX has no U.S. bonds, although it can hold them. Another example of fund names not telling the whole story.
TGBAX also hasn't had much outside of EM for a long time. But as it's called Global, it satisfies the SEC's 80% rule. It can't hold more than 20% outside the globe - which it seems to be satisfying - I didn't see any Martian or even Lunar bonds in its portfolio
Ran across this video of Hasenstab from May that gives a fair glimpse into his thinking about both interest rates (e.g. having a negative duration in US - consistent with what David wrote) and currencies.
Ran across this video of Hasenstab from May that gives a fair glimpse into his thinking about both interest rates (e.g. having a negative duration in US - consistent with what David wrote) and currencies.
Not able to access that link. Do you have another link to access it?
Ran across this video of Hasenstab from May that gives a fair glimpse into his thinking about both interest rates (e.g. having a negative duration in US - consistent with what David wrote) and currencies.
Not able to access that link. Do you have another link to access it?
The link should work (even if you're not logged in to M*), but a quick search turned up the same video at NASDAQ (but without the transcript or reader comments):
So it seems that as of September TGBAX is up to 8% Ukrainian bonds (from 1.8% prior). Morningstar also shows it at 50% cash, too.
In reviewing the 2015 annual report that just came out, I am somewhat dismayed at the 'managers commentary' ... seemed rather light on analysis and commentary and just kept harping on about "absolute return". Frankly the managers commentary seemed almost like an afterthought to them this year.
I'm still holding TGBAX but am strongly considering lightening my position due to a) 50% cash holding and b) a somewhat 'odd feeling' I had reading their managers commentary.
My PREMX holds 4.46% in Ukraine, but it's explicitly an EM bond fund. At FNMIX, another EM bond fund, Ukraine doesn't appear in the top 10 countries.
Consider this, which I just saw, dated 27th October: (re: PREMX) "...Conviction investments keep the portfolio dynamic. An overweight to Ukraine has benefited from recent debt restructurings that were in favor of bondholders. And the fund remains committed to Brazil, with investments in sovereigns and government-backed companies--in particular, Petrobras (PBR)..." LINK: http://www.morningstar.com/cover/videocenter.aspx?id=718764
Here's a less pertinent M* article. After going into it up to a point, I stopped reading. Her numbers are all screwed up. The thing is dated 02 November. She says that PREMX, (among others) is down y-t-d by -1.06%. Go to the "quote" page for the fund, and it tells you that PREMX is up, ytd by +2.60%. How reliable are her OTHER numbers in there, eh? For what it's worth, here's the link: http://news.morningstar.com/articlenet/article.aspx?id=720665
Correction: Franklin reports only 21% cash as of 9/30. Guess M* data is incorrect (an ongoing problem over there) or they're calculating 'cash' as something more than what Franklin does.
Some of the figures above mix Q3 data (portfolio composition) and 11/2/15 data (performance). The fault ... is not in our (Morning)stars, but in ourselves.
The Morningstar data in the article cited are all Q3 data. That's why you're seeing -1.06% YTD for PREMX. Go to the M* performance page for the fund, and in the Trailing Total Returns section, click on the "Quarterly" tab.
M*'s definition of cash includes bonds with maturities of up to a year. How do funds define cash? Whatever it is, I suspect it is different (much more constrained, resulting in a much lower stated cash allocation).
It is best to stick with a single data source - that way you know the same definitions and calculations apply to all the figures, and the will be reliable. Otherwise, one might wind up wondering whether PREMX (as of the end of Q3) indeed had 4.46% in Ukrane (M*'s figure) or 4.9% (T. Rowe Price's figure).
@MSF: I guess I'm a hard case. WTF? The dots aren't connecting for me re: the (admittedly often unreliable) Morningstar numbers. Bloomberg: PREMX y-t-d (after the close on 03 Nov, which was a good day) shows a positive number: +3.13%. http://www.bloomberg.com/quote/PREMX:US
The M* figures in the article are as of 9/30/2015. When one quotes YTD figures, one has to be clear on what the "date" is in "year to date". (The Nov 2 dateline of the article is the date the article was written, not necessarily the date of the data within the article.)
M* shows YTD (D = Nov 3) for PREMX of 3.14%. Likely rounding was done differently by Bloomberg; I'm not going try to figure out why the two figures differ by 0.01%. They're basically the same.
If you're looking at YTD performance that includes October and part of November, it would be nice to know what was in the portfolio that produced that performance. Unfortunately, funds report their portfolios with a 30 day lag (at least I think that's the delay), so the best you're going to get is a portfolio as of 9/30/15.
It seems to me that if that's the portfolio one is looking at, then one should also be looking at the performance that this particular portfolio achieved - that is, the returns through the same date, 9/30. Not that it makes a big difference. The portfolio is in constant flux, day by day and even minute by minute, so a snapshot still doesn't tell you what happened along the way to achieve the performance shown.
The raw data for holdings should be the same regardless of the source of you data. That is, Bloomberg, M*, the fund page, should all report exactly the same holdings for a fund on a given date.
But any analysis of the holdings (average credit rating, percent in cash, etc.) is going to vary from source to source. That's because while two sources (M*, fund page) may use the same names (e.g. percent in cash), their calculations may be different.
M* throws bonds with maturities under 1 year into the cash bucket. So if you have a portfolio that is filled entirely with bonds where half mature in six months and half mature in 10 years, M* will say that your portfolio is 50% cash, 50% bonds. My guess is that when M* analyzes the country exposure of the portfolio, it only looks at the 50% in long term bonds. And the country exposure of those bonds may be different than the country exposure of the short term bonds. So what's reported as country exposure might depend on whether you look at just the long bonds or all the bonds.
I haven't even gotten into derivatives, in part because I haven't tried thinking through how one might analyze them. Suffice to say that TGBAX plays enough games with currency exposure (it bears no relationship to the bond country exposure) that one should be able to come up with very different figures depending on how one treats derivatives.
Ideally, one should read up on how all of the numbers are calculated. (I've posted before about how M* computes average credit quality in a way that gives a lot more weight to lower graded bonds.)
If one doesn't fully understand what the numbers mean (I certainly don't), the next best thing is to stick with one source (M*, Bloomberg, some other aggregator). That way, any summary figure (e.g. average credit quality) is computed the same way for each of the funds one is comparing. M* will compute credit the same way for PREMX as for VTBIX.
That's certainly thorough. Thank you! And yet, why can't "year-to-date" simply mean that??? That is to say, starting at 01 January--- up to the moment??? I'm just grousing.
Comments
If you were looking for meek 'n' mild, it was probably a misplaced bet to begin with. If you're looking for "always wins," all bets are misplaced.
In general, he badly trails his benchmark about one quarter in four then briskly leads it in the next two. That's been the case so far in 2015: down in 2014Q4, ahead in 2015Q1 and 2015Q2.
For what that's worth,
David
I came to ask myself: would I, on my own, invest in ANY Ukrainian bonds? (not a chance, not a dime) Do I really want or need to be taking risks with the bond portion of my portfolio, to the extent that he was? I'd rather take the risks on the equity side of my portfolio.
I think the nature of some of the risks he take, are philosophically, not the kind I am prepared to take with bonds. My sense is that over the years he has gotten more "swing for the fences" in making bets. I'm not too comfortable with that type of manager in equities, let alone in bonds. But that is just me.
Just one opinion. Good luck.
So it seems the main reason to hold a bond fund is as an equity diversifier. TGBAX doesn't serve the role quite as well as a vanilla bond fund in this regard, but it does it with (as rfono observed) equity-like performance. 3 year correlation coefficients (R^2 is, well, the square of these numbers) are 0.54 for TGBAX vs. VFINX, and 0.28 for VBMFX vs. VFINX.
There's nothing else quite like Hasenstab's funds, so I think they're great as diversifiers. That includes currency.
Hasenstab almost completely decouples interest rate positions from currency positions. He can be long foreign bonds (for better yields) while short foreign currency. That's part of the appeal of his management style. Right now, TGBAX is 78% long in US dollars (more than double the benchmark Citigroup World Gov. Bond Index), and short Euros (24%) and Yen (36%).
Hasenstab is one of them.
prinx
What time frame are you thinking about when talking about performance "at this time"? YTD? 1 year, 1 month?
If some of these other managers are doing even worse at this time, would staying with them also be misplaced loyalty?
TGBAX is down 1.75% YTD. LSBDX (Fuss) is down 3.62%. His protege is doing even worse - down 6.04% at EVBIX. (Data from M*, as of Aug 7th.)
It seems to me that loyalty is not misplaced if long term managers have demonstrated superior performance and discipline over time, and circumstances have not changed. Are these managers doing worse than would be expected for their various fund categories?
If not, then we're not talking about disliking the manager, but disliking categories. So perhaps we should be discussing asset allocation, not manager performance.
More to do with this investment area (global bonds) in general and not favoring one manager over another manager.
Edit note: I incorrectly noted a 20% equity position in the past with LSBDX. @msf has corrected this with his below report. Thank you, msf.
Fidelity via M* composition for LSBDX dated 5-31-15
The fund has tended to add a little equity in recent years, but nothing over 10%. A higher equity position must have been a long time ago. Here's what I found in the (semi) annual reports over the past few years:
Common + Preferred Stock by report date:
3/31/15: 6.4% + 1.8%
9/30/14: 5.1% + 1.9%
3/31/14: 6.8% + 2.0%
9/30/13: 5.8% + 3.6%
3/31/13: 5.4% + 2.9%
9/30/12: 4.9% + 2.5%
3/31/12: 5.4% + 2.6%
9/30/11: 4.8% + 2.6%
3/31/11: 2.6% + 3.3%
9/30/10: 2.3% + 2.6%
3/31/10: 1.5% + 2.5%
9/30/09: 0.7% + 1.4%
3/31/09: 0.0% + 0.1%
9/30/08: 0.7% + 1.2%
3/31/08: 1.1% + 1.7%
In the world of investing, investors often falls victim to the most common denominator of reward over shorter and shorter periods of time when what makes bond manager's great may have more to do with the way they manage risk over longer and longer periods of time.
I have considered pairing a fund like TGBAX with an uncorrelated world equity fund.
Pairing TGBAX with Vanguard's Global Minimum Volatility Fund (VTWSX) might be a conservative way to gain that "balanced bond/equity" exposure to world investments.
Ran across this video of Hasenstab from May that gives a fair glimpse into his thinking about both interest rates (e.g. having a negative duration in US - consistent with what David wrote) and currencies.
Also from today (Aug 12): Bloomberg video (and written report): Ukraine Bonds Jump Before Templeton Talks Amid Debt Deal Bets.
Do you have another link to access it?
http://www.nasdaq.com/video/hasenstab--flexibility-key-in-global-bonds-518842820
So it seems that as of September TGBAX is up to 8% Ukrainian bonds (from 1.8% prior). Morningstar also shows it at 50% cash, too.
In reviewing the 2015 annual report that just came out, I am somewhat dismayed at the 'managers commentary' ... seemed rather light on analysis and commentary and just kept harping on about "absolute return". Frankly the managers commentary seemed almost like an afterthought to them this year.
I'm still holding TGBAX but am strongly considering lightening my position due to a) 50% cash holding and b) a somewhat 'odd feeling' I had reading their managers commentary.
MAINX ytd +0.45% 1-yr: -0.95% 3-yr: +1.31% (too young for 5-year number.)
PRSNX ytd +0.60% 1-yr: -1.14% 3-yr: +2.18% 5-yr: +3.48%
My PREMX holds 4.46% in Ukraine, but it's explicitly an EM bond fund.
At FNMIX, another EM bond fund, Ukraine doesn't appear in the top 10 countries.
Consider this, which I just saw, dated 27th October: (re: PREMX)
"...Conviction investments keep the portfolio dynamic. An overweight to Ukraine has benefited from recent debt restructurings that were in favor of bondholders. And the fund remains committed to Brazil, with investments in sovereigns and government-backed companies--in particular, Petrobras (PBR)..."
LINK: http://www.morningstar.com/cover/videocenter.aspx?id=718764
Here's a less pertinent M* article. After going into it up to a point, I stopped reading. Her numbers are all screwed up. The thing is dated 02 November. She says that PREMX, (among others) is down y-t-d by -1.06%. Go to the "quote" page for the fund, and it tells you that PREMX is up, ytd by +2.60%. How reliable are her OTHER numbers in there, eh?
For what it's worth, here's the link:
http://news.morningstar.com/articlenet/article.aspx?id=720665
The fault ... is not in our (Morning)stars, but in ourselves.
The Morningstar data in the article cited are all Q3 data. That's why you're seeing -1.06% YTD for PREMX. Go to the M* performance page for the fund, and in the Trailing Total Returns section, click on the "Quarterly" tab.
M*'s definition of cash includes bonds with maturities of up to a year. How do funds define cash? Whatever it is, I suspect it is different (much more constrained, resulting in a much lower stated cash allocation).
It is best to stick with a single data source - that way you know the same definitions and calculations apply to all the figures, and the will be reliable. Otherwise, one might wind up wondering whether PREMX (as of the end of Q3) indeed had 4.46% in Ukrane (M*'s figure) or 4.9% (T. Rowe Price's figure).
Bloomberg: PREMX y-t-d (after the close on 03 Nov, which was a good day) shows a positive number: +3.13%.
http://www.bloomberg.com/quote/PREMX:US
The M* figures in the article are as of 9/30/2015. When one quotes YTD figures, one has to be clear on what the "date" is in "year to date". (The Nov 2 dateline of the article is the date the article was written, not necessarily the date of the data within the article.)
M* shows YTD (D = Nov 3) for PREMX of 3.14%. Likely rounding was done differently by Bloomberg; I'm not going try to figure out why the two figures differ by 0.01%. They're basically the same.
If you're looking at YTD performance that includes October and part of November, it would be nice to know what was in the portfolio that produced that performance. Unfortunately, funds report their portfolios with a 30 day lag (at least I think that's the delay), so the best you're going to get is a portfolio as of 9/30/15.
It seems to me that if that's the portfolio one is looking at, then one should also be looking at the performance that this particular portfolio achieved - that is, the returns through the same date, 9/30. Not that it makes a big difference. The portfolio is in constant flux, day by day and even minute by minute, so a snapshot still doesn't tell you what happened along the way to achieve the performance shown.
The raw data for holdings should be the same regardless of the source of you data. That is, Bloomberg, M*, the fund page, should all report exactly the same holdings for a fund on a given date.
But any analysis of the holdings (average credit rating, percent in cash, etc.) is going to vary from source to source. That's because while two sources (M*, fund page) may use the same names (e.g. percent in cash), their calculations may be different.
M* throws bonds with maturities under 1 year into the cash bucket. So if you have a portfolio that is filled entirely with bonds where half mature in six months and half mature in 10 years, M* will say that your portfolio is 50% cash, 50% bonds. My guess is that when M* analyzes the country exposure of the portfolio, it only looks at the 50% in long term bonds. And the country exposure of those bonds may be different than the country exposure of the short term bonds. So what's reported as country exposure might depend on whether you look at just the long bonds or all the bonds.
I haven't even gotten into derivatives, in part because I haven't tried thinking through how one might analyze them. Suffice to say that TGBAX plays enough games with currency exposure (it bears no relationship to the bond country exposure) that one should be able to come up with very different figures depending on how one treats derivatives.
Ideally, one should read up on how all of the numbers are calculated. (I've posted before about how M* computes average credit quality in a way that gives a lot more weight to lower graded bonds.)
If one doesn't fully understand what the numbers mean (I certainly don't), the next best thing is to stick with one source (M*, Bloomberg, some other aggregator). That way, any summary figure (e.g. average credit quality) is computed the same way for each of the funds one is comparing. M* will compute credit the same way for PREMX as for VTBIX.