Critique Fixed Income Strategy - GNMA Dominance

Have a question about your personal investments? No matter how simple or complex, you can ask it here.
Post Reply
Buffetologist
Posts: 299
Joined: Sun Aug 22, 2010 8:58 am

Critique Fixed Income Strategy - GNMA Dominance

Post by Buffetologist » Tue Apr 26, 2016 7:11 am

Looking for some insightful critique beyond mantras like "Keep it simple" or that I'm "overthinking". I like overthinking.

So I'm looking at my fixed income portfolio. Can't list the whole thing, but just looking at my dominant options I keep coming up with the notion that the Fidelity GNMA fund dominates any other combination of fungible assets. By dominating, I mean that comparing the SEC yield and duration of the GNMA fund with a weighted barbell of any other fungible option yields something with more credit risk and/or duration risk that I don't think I'm compensated for. The only exception is the Fidelity Mortgage Securities Fund which provides an additional 38 basis points with the roughly the same duration as the GNMA to go from AAA to A. Not sure if that's enough compensation for the additional credit risk. There are some longer duration Fidelity options that might seem to provide sufficient compensation, but when barbelled with my lower duration options to equate the duration, are always inferior to the GNMA and come with additional credit risk.

We're high income. Average age 53. I tend to multiply my tax advantaged 401Ks and IRA by my expected tax rate in retirement and then form a joint portfolio with the taxable stuff. We're targeting 65% equities and 35% fixed income which include 4% Ibonds which I keep as a separate category and have maxed out. So the 31% is divided between cash and fixed income with some duration. I'll call that my fixed income portfolio. I then monitor my yield and duration to form a portfolio. I try to maximize the after-tax yield subject to a weighted duration of about 2.5 to 3 years. I buy my maximum amount of EE bonds every year and that provides long duration with downside protection. I also have an IRA which is mostly committed to credit union CDs well above the treasury yield curve. I should also say that more than half of the entire portfolio (stock+fixed income) is in taxable. My expected tax rate in retirement is 31.71%, my tax rate on qualified dividends and capital gains is 28% (includes state), so it really doesn't matter a whole lot where my stocks are placed. However, given my current tax rate on interest is 47.64% the benefit of tax advantage on ordinary income is huge. The 47.64% comes from 39.6% tax rate plus 1.188% itemized deduction phaseout, plus 3.8% investment tax, plus 3.05% state tax (5.15% minus the federal deduction).

Fungible options I have in taxable with their after tax yields are
Ally Savings 0.52%
Vanguard Limited Term tax exempt admiral 0.93% 2.5 years AA (dominates after tax of short-term investment grade admiral and short-term corporate index admiral, but far inferior to my tax advantage options).
Vanguard Intermediate tax exempt admiral 1.41%, 4.8 years AA (Only here for illustration. I dumped this recently because the yield got too low. This is totally dominated by my tax advantaged options. It's even dominated by my mortgage which I am loathe to prepay. When I bought this it was 2.15%)
Mortgage prepayment 1.63% (2.75% mortgage, 82 payments left, average duration 3.56, but of course prepayments get applied to the long end first)

Fungible Tax Advantage Options - using 30-day SEC yields, since i have already discounted the principal based on my expected tax rate
MFS Limited Maturity R5 1.27%, 1.43 years, A
Fidelity Limited Term Bond 1.93%, 2.58 years, A
Fidelity GNMA, 2.13%, 3.02 years, AAA
Fidelity Mortgage Securities 2.51%, 2.98 years, A
Fidelity Investment Grade 2.91%, 5.67 years, A
Fidelity Total Bond 3.3%, 5.14 years, BBB
Fidelity Corporate Bond fund 3.52%, 6.81 years, BBB

There is also a MetLife Stable Value with a 3% guarantee but because of the surrender charges you can only remove 20% every year without penalty, so in effective it asymptotically has a 5 year duration. The credit is based on MetLife's claims paying ability which is an undiversified obligation of a single creditor who won't succumb to attempts by the fed to force it to manage it's risk, so let's generously call that single A. Not crazy about this option.

I can not find a current barbell combination of (alpha)*InvestmentA + (1-alpha)*InvestmentB that comes anywhere near Fidelity GNMA in terms of yield, duration and credit risk. However, I am aware that I am being compensated in the GNMA for the negative convexity of mortgages. I don't know how to factor that into the selection.

My fixed income portfolio is currently 41% GNMA, (calculated assuming that my own mortgage is really a -14% short position on a riskless security).
My current duration is a little lower than I'd like (2.47 years), but no other option for moving it higher seems as good as more mortgages.

I'm trying to figure out if it's worth sacrificing some yield to get some more credit diversity even though doing so will entail more credit and duration risk. I'm also trying to assess whether the 38 basis points provided by the Mortgage Securities Fund to drop from AAA to A is enough compensation, or if the 78 basis points to do the Investment Grade is enough compensation to extend the duration and lower the credit. We prefer our risk to be in equities, so I'm also not crazy about the BBB options, but am willing to rethink that too. Using the Morningstar approach of AAA = 0, AA = 1, A = 2, BBB = 3, etc, the current risk of the fixed income portfolio is currently 0.18. I would be comfortable with a little more credit risk.

Basically, I would use cash in the Ally Savings to buy VanguardIdx500 and move tax advantage funds from Fidelity Spartan 500 into more fixed income options.

Any comments on the overall approach? How do you weight negative convexity? Do the longer duration Fidelity funds seem at all attractive despite the credit risk?

Appreciate any insight especially of the mathematical variety. Thanks in advance.

FillorKill
Posts: 1007
Joined: Sat Aug 06, 2011 7:01 am

Re: Critique Fixed Income Strategy - GNMA Dominance

Post by FillorKill » Tue Apr 26, 2016 8:41 am

Buffetologist wrote:However, I am aware that I am being compensated in the GNMA for the negative convexity of mortgages. I don't know how to factor that into the selection.

Consider it compensation for the increased duration volatility that the negative convexity creates.

Buffetologist
Posts: 299
Joined: Sun Aug 22, 2010 8:58 am

Re: Critique Fixed Income Strategy - GNMA Dominance

Post by Buffetologist » Tue Apr 26, 2016 8:48 am

FillorKill wrote:
Buffetologist wrote:However, I am aware that I am being compensated in the GNMA for the negative convexity of mortgages. I don't know how to factor that into the selection.

Consider it compensation for the increased duration volatility that the negative convexity creates.



Thanks for your reply.

Interesting concept - duration volatility. Hadn't considered that viewpoint.

So how would you factor it into your analysis? How much would you discount the yield on GNMAs to be comparable with less volatile durations?

Valuethinker
Posts: 32925
Joined: Fri May 11, 2007 11:07 am

Re: Critique Fixed Income Strategy - GNMA Dominance

Post by Valuethinker » Tue Apr 26, 2016 10:16 am

Negative convexity means the duration of the bond changes with moves in interest rates.

It goes up when interest rates rise, and down when they fall.

You'd need some pretty complex software and lots of data to properly adjust for that. Fund managers do, and they get it wrong.

It's unlikely there is some unarbitraged market anomaly there. Basically the return you get is a function of the risk you take.

User avatar
stratton
Posts: 11080
Joined: Sun Mar 04, 2007 5:05 pm
Location: Puget Sound

Re: Critique Fixed Income Strategy - GNMA Dominance

Post by stratton » Tue Apr 26, 2016 10:58 am

Considering expense ratio comes out of yield... Was that accounted for in the calculations?

Paul
...and then Buffy staked Edward. The end.

tyc
Posts: 109
Joined: Wed Mar 30, 2016 1:36 pm

Re: Critique Fixed Income Strategy - GNMA Dominance

Post by tyc » Tue Apr 26, 2016 12:56 pm

Guess that's why I don't like fixed income investments. There are too many risk factors and very low returns. However, fixed income works great for portfolio diversification.

Most analysis are done before expenses. Even if you have all the data you need, you'll still wouldn't be able to correctly neutralized all your risks because transaction costs would eat into your portfolio returns. Keep you analysis simple and focus only on duration and yield. Then uses expense ratio to justified if you're paying too much for your exposures. A simple correlation between different funds return will give you a simple idea if the funds are similar are not.

Buffetologist
Posts: 299
Joined: Sun Aug 22, 2010 8:58 am

Re: Critique Fixed Income Strategy - GNMA Dominance

Post by Buffetologist » Wed Apr 27, 2016 7:59 am

Everything I've quoted for funds is the SEC yield minus, for taxable stuff, my tax cost.

Valuethinker, are you saying that there is no reason for me to be in GNMA vs treasuries and that the extra yield to compensate for the negative convexity isn't really compensating me?

FoolStreet
Posts: 425
Joined: Fri Sep 07, 2012 12:18 am

Re: Critique Fixed Income Strategy - GNMA Dominance

Post by FoolStreet » Wed Apr 27, 2016 8:22 am

Buffetologist wrote:Looking for some insightful critique beyond mantras like "Keep it simple" or that I'm "overthinking". I like overthinking.

So I'm looking at my fixed income portfolio. Can't list the whole thing, but just looking at my dominant options I keep coming up with the notion that the Fidelity GNMA fund dominates any other combination of fungible assets. By dominating, I mean that comparing the SEC yield and duration of the GNMA fund with a weighted barbell of any other fungible option yields something with more credit risk and/or duration risk that I don't think I'm compensated for. The only exception is the Fidelity Mortgage Securities Fund which provides an additional 38 basis points with the roughly the same duration as the GNMA to go from AAA to A. Not sure if that's enough compensation for the additional credit risk. There are some longer duration Fidelity options that might seem to provide sufficient compensation, but when barbelled with my lower duration options to equate the duration, are always inferior to the GNMA and come with additional credit risk.

We're high income. Average age 53. I tend to multiply my tax advantaged 401Ks and IRA by my expected tax rate in retirement and then form a joint portfolio with the taxable stuff. We're targeting 65% equities and 35% fixed income which include 4% Ibonds which I keep as a separate category and have maxed out. So the 31% is divided between cash and fixed income with some duration. I'll call that my fixed income portfolio. I then monitor my yield and duration to form a portfolio. I try to maximize the after-tax yield subject to a weighted duration of about 2.5 to 3 years. I buy my maximum amount of EE bonds every year and that provides long duration with downside protection. I also have an IRA which is mostly committed to credit union CDs well above the treasury yield curve. I should also say that more than half of the entire portfolio (stock+fixed income) is in taxable. My expected tax rate in retirement is 31.71%, my tax rate on qualified dividends and capital gains is 28% (includes state), so it really doesn't matter a whole lot where my stocks are placed. However, given my current tax rate on interest is 47.64% the benefit of tax advantage on ordinary income is huge. The 47.64% comes from 39.6% tax rate plus 1.188% itemized deduction phaseout, plus 3.8% investment tax, plus 3.05% state tax (5.15% minus the federal deduction).

Fungible options I have in taxable with their after tax yields are
Ally Savings 0.52%
Vanguard Limited Term tax exempt admiral 0.93% 2.5 years AA (dominates after tax of short-term investment grade admiral and short-term corporate index admiral, but far inferior to my tax advantage options).
Vanguard Intermediate tax exempt admiral 1.41%, 4.8 years AA (Only here for illustration. I dumped this recently because the yield got too low. This is totally dominated by my tax advantaged options. It's even dominated by my mortgage which I am loathe to prepay. When I bought this it was 2.15%)
Mortgage prepayment 1.63% (2.75% mortgage, 82 payments left, average duration 3.56, but of course prepayments get applied to the long end first)

Fungible Tax Advantage Options - using 30-day SEC yields, since i have already discounted the principal based on my expected tax rate
MFS Limited Maturity R5 1.27%, 1.43 years, A
Fidelity Limited Term Bond 1.93%, 2.58 years, A
Fidelity GNMA, 2.13%, 3.02 years, AAA
Fidelity Mortgage Securities 2.51%, 2.98 years, A
Fidelity Investment Grade 2.91%, 5.67 years, A
Fidelity Total Bond 3.3%, 5.14 years, BBB
Fidelity Corporate Bond fund 3.52%, 6.81 years, BBB

There is also a MetLife Stable Value with a 3% guarantee but because of the surrender charges you can only remove 20% every year without penalty, so in effective it asymptotically has a 5 year duration. The credit is based on MetLife's claims paying ability which is an undiversified obligation of a single creditor who won't succumb to attempts by the fed to force it to manage it's risk, so let's generously call that single A. Not crazy about this option.

I can not find a current barbell combination of (alpha)*InvestmentA + (1-alpha)*InvestmentB that comes anywhere near Fidelity GNMA in terms of yield, duration and credit risk. However, I am aware that I am being compensated in the GNMA for the negative convexity of mortgages. I don't know how to factor that into the selection.

My fixed income portfolio is currently 41% GNMA, (calculated assuming that my own mortgage is really a -14% short position on a riskless security).
My current duration is a little lower than I'd like (2.47 years), but no other option for moving it higher seems as good as more mortgages.

I'm trying to figure out if it's worth sacrificing some yield to get some more credit diversity even though doing so will entail more credit and duration risk. I'm also trying to assess whether the 38 basis points provided by the Mortgage Securities Fund to drop from AAA to A is enough compensation, or if the 78 basis points to do the Investment Grade is enough compensation to extend the duration and lower the credit. We prefer our risk to be in equities, so I'm also not crazy about the BBB options, but am willing to rethink that too. Using the Morningstar approach of AAA = 0, AA = 1, A = 2, BBB = 3, etc, the current risk of the fixed income portfolio is currently 0.18. I would be comfortable with a little more credit risk.

Basically, I would use cash in the Ally Savings to buy VanguardIdx500 and move tax advantage funds from Fidelity Spartan 500 into more fixed income options.

Any comments on the overall approach? How do you weight negative convexity? Do the longer duration Fidelity funds seem at all attractive despite the credit risk?

Appreciate any insight especially of the mathematical variety. Thanks in advance.



Vanguard shows GNMA fund SEC of 2.37 and negative con exits as mentioned previously. Why not use stable value fund?

FillorKill
Posts: 1007
Joined: Sat Aug 06, 2011 7:01 am

Re: Critique Fixed Income Strategy - GNMA Dominance

Post by FillorKill » Wed Apr 27, 2016 8:47 am

Buffetologist wrote:...are you saying that there is no reason for me to be in GNMA vs treasuries and that the extra yield to compensate for the negative convexity isn't really compensating me?

Ex-ante it is most reasonable to assume that the extra yield is sufficient compensation for the additional risks. Attempting to make an active decision based on your assessment of the pricing of that risk is what I believe Valuethinker was warning you off of. Understanding the risks you assume and not becoming overly concentrated in any one sector of the debt market are obviously advisable (in no way do you appear to be pursuing such a path).

I'm not very familiar with Fidelity funds so I'll use VG as a point of comparison. I am not aware of a fixed income VG fund that has higher 15Y Sharpe/Sortino ratios than the GNMA fund (VFIJX) 1.19/2.09 respectively. Those are very impressive and those risks were handsomely rewarded over the period. The 15Y Sharpe/Sortino for the Intermediate Treasury Fund (VFIUX) are .77/1.32. No where near as impressive. As an additional data point the 15Y Sharpe/Sortino for Total Bond (VBTLX) were: .94/1.59

The VG GNMA fund more than rewarded the additional risks over the period IMO.

EDIT: Here are the 15Y Sharpe/Sortino of the Fidelity GNMA fund (FGMNX) 1.23/2.33

User avatar
abuss368
Posts: 11568
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!

Re: Critique Fixed Income Strategy - GNMA Dominance

Post by abuss368 » Wed Apr 27, 2016 9:22 am

In my opinion any low cost and diversified short or intermediate term bond fund will provide safety and income to an investment portfolio.
John C. Bogle: "You simply do not need to put your money into 8 different mutual funds!" | | Disclosure: Three Fund Portfolio + U.S. & International REITs

Valuethinker
Posts: 32925
Joined: Fri May 11, 2007 11:07 am

Re: Critique Fixed Income Strategy - GNMA Dominance

Post by Valuethinker » Wed Apr 27, 2016 10:58 am

Buffetologist wrote:Everything I've quoted for funds is the SEC yield minus, for taxable stuff, my tax cost.

Valuethinker, are you saying that there is no reason for me to be in GNMA vs treasuries and that the extra yield to compensate for the negative convexity isn't really compensating me?


GNMA bonds pay an extra yield (the Option Adjusted Spread) because the borrower has an option to repay early/ extend the mortgage-- that is what creates the negative convexity.

That spread moves in and out with market conditions. You can't time it.

You might get an extra return from holding GNMA, because the worser cases never happened-- the Fed didn't raise rates suddenly (extension risk) or mortgage holders didn't refinance ahead of expectations.

I would tend to hold Treasuries because they are a "purer" play on interest rate risk. So I am pretty clear what risks I am taking. That's fuzzier with GNMA bonds (any US Mortgage Backed Security fund).

However over the last 15 years you would have been better off in GNMAs, I believe.

Buffetologist
Posts: 299
Joined: Sun Aug 22, 2010 8:58 am

Re: Critique Fixed Income Strategy - GNMA Dominance

Post by Buffetologist » Wed Apr 27, 2016 11:09 am

FoolStreet wrote:
Vanguard shows GNMA fund SEC of 2.37 and negative con exits as mentioned previously. Why not use stable value fund?



My 401K is with Fidelity, so Vanguard is not an option for those funds.

Perhaps I'm misunderstanding, but stable value is guaranteed by the claims paying ability of MetLife. Unlike GNMA where at least the underlying securities are backed by the full faith and credit of the US Government, stable value is a completely undiversified obligation of a single creditor. Frankly, I don't trust the health of the entire life insurance industry. They have all of these existing guaranteed annuities and cash value policies that have minimum guarantees at rates for which they really have no mechanism to earn without taking on substantial risk that lowers their credit rating. I'm thinking dog droppings wrapped in cat droppings (to sort of more tastefully quote The Big Short movie). Additionally, they have the regulators trying to prevent them from taking undue risk so that they cease to be a systematically important company. Not taking these risks causes them to lose money because of existing obligations, also lowering their credit rating. May not end well and I'm uncomfortable committing substantial funds to this potential bubble.

Furthermore, the stable value fund has liquidity issues and surrender charges. Basically, you can only move out 20% per year without paying a 4% surrender charge. If I weight the duration by taking out 20% per year, to calculate the average maturity, I get a geometric series 0.2*1+ (0.2)(0.8)*2+ (0.2)*(0.8)(.8)*3 + .... = 5 in the limit. So it's basically a 5 year duration, 3% return based on the claims paying ability of MetLife.

Since I plan to work for the foreseeable future, this does not seem like a fungible option.

Buffetologist
Posts: 299
Joined: Sun Aug 22, 2010 8:58 am

Re: Critique Fixed Income Strategy - GNMA Dominance

Post by Buffetologist » Wed Apr 27, 2016 11:19 am

FillorKill wrote:Ex-ante it is most reasonable to assume that the extra yield is sufficient compensation for the additional risks. Attempting to make an active decision based on your assessment of the pricing of that risk is what I believe Valuethinker was warning you off of. Understanding the risks you assume and not becoming overly concentrated in any one sector of the debt market are obviously advisable (in no way do you appear to be pursuing such a path).

I'm not very familiar with Fidelity funds so I'll use VG as a point of comparison. I am not aware of a fixed income VG fund that has higher 15Y Sharpe/Sortino ratios than the GNMA fund (VFIJX) 1.19/2.09 respectively. Those are very impressive and those risks were handsomely rewarded over the period. The 15Y Sharpe/Sortino for the Intermediate Treasury Fund (VFIUX) are .77/1.32. No where near as impressive. As an additional data point the 15Y Sharpe/Sortino for Total Bond (VBTLX) were: .94/1.59

The VG GNMA fund more than rewarded the additional risks over the period IMO.

EDIT: Here are the 15Y Sharpe/Sortino of the Fidelity GNMA fund (FGMNX) 1.23/2.33


Thanks for your analysis FillorKill (and yours too ValueThinker).

What's the maximum of your fixed income (exclusive of Ibonds) would you put in the GNMA or in Mortgages to keep from being overly concentrated?

Does it make sense to view my own mortgage as a short position on the same sector? If rates go down to 2.5 no points, no closing 10 year, I would refinance myself. In that case I'm 41%-14% = 27% in mortgages.

Maybe I should add some Fidelity Investment Grade just to get some credit diversity. I think that's my next most attractive option.

FillorKill
Posts: 1007
Joined: Sat Aug 06, 2011 7:01 am

Re: Critique Fixed Income Strategy - GNMA Dominance

Post by FillorKill » Wed Apr 27, 2016 1:44 pm

Buffetologist wrote:What's the maximum of your fixed income (exclusive of Ibonds) would you put in the GNMA or in Mortgages to keep from being overly concentrated?

Does it make sense to view my own mortgage as a short position on the same sector? If rates go down to 2.5 no points, no closing 10 year, I would refinance myself. In that case I'm 41%-14% = 27% in mortgages.

Maybe I should add some Fidelity Investment Grade just to get some credit diversity. I think that's my next most attractive option.


No problem. MBS constitutes a large portion of the US investment grade bond market. While not a fully comprehensive representation of that market, total bond has an allocation of around 22% to MBS. Not including your own mortgage you're approaching double that. It would be difficult for me to suggest you should add to your overweight. You could come to regret a significant overweight followed by prolonged or pronounced underperformance. Adding some of the investment grade fund for the diversification sounds like good idea.

I don't think counting your mortgage as a 100% short position against the GNMA position is quite accurate. Your potential refinancing has a mitigating effect on the downside (contraction risk) of a continued decline in rates on your GNMA allocation but there is a limit to that. On the other side of your refinance is a smaller total mortgage amount which is a reduction in the short (a proportional increase in the GNMA allocation). Rising rates present the potential for extension risk at which point you would hold your existing mortgage and the benefit would mitigate the downside experienced in your GNMA allocation. Either way, counting your mortgage as a 100% short, all else equal, leaves you in the position of a relatively increasing proportion in GNMA as you pay down the short.

IMO your targeted duration is a little on the short side. Adding some Term risk to get your targeted duration closer to the intermediate range may be worth consideration. As is usually the case the yield curve is steeper from 3-7 than 1-3. I'd consider a little more term risk in the portfolio.
https://www.treasury.gov/resource-cente ... ation.aspx

Buffetologist
Posts: 299
Joined: Sun Aug 22, 2010 8:58 am

Re: Critique Fixed Income Strategy - GNMA Dominance

Post by Buffetologist » Wed Apr 27, 2016 3:27 pm

FillorKill - Once again, thank you for your insight. It's quite illuminating. I like the way you think.

So basically, I should increase duration without increasing mortgages, and I can probably afford a little more credit risk. At 0.18 I'm completely dominated by full-faith and credit of the USA.

I think I'll have to increase duration with some of the longer duration Fidelity funds that also add more credit risk.

EDIT: Forgive my math but from your link, as of today, the slope is from 1-3 is 0.215 and the slope from 3-7 is 0.17. Am I calculating it wrong?

(Yield 3-Yield 1)/(3-1) = (1.04-0.61)/2 = 0.215
(Yield 7-Yield 3)/(7-3) = (1.72-1.04)/4 = 0.17

User avatar
grabiner
Advisory Board
Posts: 20695
Joined: Tue Feb 20, 2007 11:58 pm
Location: Columbia, MD

Re: Critique Fixed Income Strategy - GNMA Dominance

Post by grabiner » Wed Apr 27, 2016 7:46 pm

Buffetologist wrote:
FillorKill wrote:
Buffetologist wrote:However, I am aware that I am being compensated in the GNMA for the negative convexity of mortgages. I don't know how to factor that into the selection.

Consider it compensation for the increased duration volatility that the negative convexity creates.



Thanks for your reply.

Interesting concept - duration volatility. Hadn't considered that viewpoint.

So how would you factor it into your analysis? How much would you discount the yield on GNMAs to be comparable with less volatile durations?


I would trust the market here. GNMAs have higher yields than intermediate-term Treasuries, even though both have zero principal risk; presumably, investors are demanding those higher yields as compensation for the prepayment risk.
David Grabiner

FoolStreet
Posts: 425
Joined: Fri Sep 07, 2012 12:18 am

Re: Critique Fixed Income Strategy - GNMA Dominance

Post by FoolStreet » Wed Apr 27, 2016 8:37 pm

Buffetologist wrote:
FoolStreet wrote:
Vanguard shows GNMA fund SEC of 2.37 and negative con exits as mentioned previously. Why not use stable value fund?



My 401K is with Fidelity, so Vanguard is not an option for those funds.

Perhaps I'm misunderstanding, but stable value is guaranteed by the claims paying ability of MetLife. Unlike GNMA where at least the underlying securities are backed by the full faith and credit of the US Government, stable value is a completely undiversified obligation of a single creditor. Frankly, I don't trust the health of the entire life insurance industry. They have all of these existing guaranteed annuities and cash value policies that have minimum guarantees at rates for which they really have no mechanism to earn without taking on substantial risk that lowers their credit rating. I'm thinking dog droppings wrapped in cat droppings (to sort of more tastefully quote The Big Short movie). Additionally, they have the regulators trying to prevent them from taking undue risk so that they cease to be a systematically important company. Not taking these risks causes them to lose money because of existing obligations, also lowering their credit rating. May not end well and I'm uncomfortable committing substantial funds to this potential bubble.

Furthermore, the stable value fund has liquidity issues and surrender charges. Basically, you can only move out 20% per year without paying a 4% surrender charge. If I weight the duration by taking out 20% per year, to calculate the average maturity, I get a geometric series 0.2*1+ (0.2)(0.8)*2+ (0.2)*(0.8)(.8)*3 + .... = 5 in the limit. So it's basically a 5 year duration, 3% return based on the claims paying ability of MetLife.

Since I plan to work for the foreseeable future, this does not seem like a fungible option.


Is the single provider an assumption or fact? Check the prospectus to see if there are numerous diversified insurers and what is the credit qyuality of them. It may not be too bad. A 5yr duration for long term doesn't sound bad. But no pressure. I guess I'm being devils advocate due to the higher rate and no (?) risk of NAV loss.

Buffetologist
Posts: 299
Joined: Sun Aug 22, 2010 8:58 am

Re: Critique Fixed Income Strategy - GNMA Dominance

Post by Buffetologist » Wed Apr 27, 2016 9:18 pm

The Stable Value Summary under Risk says

"The fund is backed by a diversified portfolio of fixed-income assets held in the general account of the issuer. Guarantees are subject to the claims paying ability of the issuer. Restrictions or fees may apply to exchanges or withdrawals. The Contracts provide for the payment of certain withdrawals and exchanges at book value during the terms of the Contracts. In order to maintain the Contract issuers' promise to pay such withdrawals and exchanges at book value, the Contracts subject the fund and its participants to certain restrictions. For example, withdrawals prompted by certain events (e.g., layoffs, early retirement windows, spin-offs, sale of a division, facility closings, plan terminations, partial plan terminations, changes in laws or regulations) may be paid at the market value of the fund's securities, which may be less than your book value balance or may restrict withdrawals in these events.

This annuity contract contains withdrawal charges and terms for keeping it in force. No restrictions or additional fees, including surrender charges, will apply to the amount you withdraw as a result of certain life events. Please see your certificate or contact Fidelity for more information.

Partial withdrawals or exchanges of up to 20.00% of your account balance may be made in any contract year without incurring contract surrender charges. Withdrawals or exchanges in excess of 20.00% may be subject to a surrender charge of up to 4.00% of the amount of the excess withdrawal or exchange. Please see your certificate or contact Fidelity for more information."

Am I misunderstanding what they say by "Guarantees are subject to the claims paying ability of the issuer."?

FoolStreet
Posts: 425
Joined: Fri Sep 07, 2012 12:18 am

Re: Critique Fixed Income Strategy - GNMA Dominance

Post by FoolStreet » Wed Apr 27, 2016 11:10 pm

Buffetologist wrote:The Stable Value Summary under Risk says

"The fund is backed by a diversified portfolio of fixed-income assets held in the general account of the issuer. Guarantees are subject to the claims paying ability of the issuer. Restrictions or fees may apply to exchanges or withdrawals. The Contracts provide for the payment of certain withdrawals and exchanges at book value during the terms of the Contracts. In order to maintain the Contract issuers' promise to pay such withdrawals and exchanges at book value, the Contracts subject the fund and its participants to certain restrictions. For example, withdrawals prompted by certain events (e.g., layoffs, early retirement windows, spin-offs, sale of a division, facility closings, plan terminations, partial plan terminations, changes in laws or regulations) may be paid at the market value of the fund's securities, which may be less than your book value balance or may restrict withdrawals in these events.

This annuity contract contains withdrawal charges and terms for keeping it in force. No restrictions or additional fees, including surrender charges, will apply to the amount you withdraw as a result of certain life events. Please see your certificate or contact Fidelity for more information.

Partial withdrawals or exchanges of up to 20.00% of your account balance may be made in any contract year without incurring contract surrender charges. Withdrawals or exchanges in excess of 20.00% may be subject to a surrender charge of up to 4.00% of the amount of the excess withdrawal or exchange. Please see your certificate or contact Fidelity for more information."

Am I misunderstanding what they say by "Guarantees are subject to the claims paying ability of the issuer."?


I think you are misunderstanding why I am asking. Other stable value funds are comprised of multiple insurer contracts.

Here is Vanguard's: The fund invests primarily in synthetic investment contracts backed by high-credit-quality fixed income investments and traditional investments issued by insurance companies and banks.

Here is what mine says: The fund manager seeks to control risk by using a diversified portfolio of generally high-quality investments and enters into wrap contracts with highly rated entities. In general, bond prices rise when interest rates fall, and vice versa. The effect is usually more pronounced with longer-term securities. Bonds also face credit risk, the possibility that bond issuers may be unable to make payments. The fund uses derivatives to obtain synthetic exposure and for risk management purposes, which involves risks such as credit risk. For wrapped bonds, the wrap contracts allow price changes in the bonds to be smoothed into the crediting rate over time. There is credit risk associated with the wrap issuers.


Therefore, I am used to seeing the risk spread across multiple issuers, presumably of high quality. In my case, there is a list of issuers and their high credit quality.

If after you read your prospectus details, you find that you are dependent on only one issuer of dubious credit, then of course I agree with you: steer clear!

FillorKill
Posts: 1007
Joined: Sat Aug 06, 2011 7:01 am

Re: Critique Fixed Income Strategy - GNMA Dominance

Post by FillorKill » Thu Apr 28, 2016 10:26 am

Buffetologist wrote:I think I'll have to increase duration with some of the longer duration Fidelity funds that also add more credit risk.

EDIT: Forgive my math but from your link, as of today, the slope is from 1-3 is 0.215 and the slope from 3-7 is 0.17.

Your algebra may be better than my statement :D I last looked earlier in the month at which point the steepest segment among the 1-3, 3-5, 5-7 ranges was 3-5, followed by 5-7 and then 1-3. Looks that that arrangement has shifted a bit.

Not to obscure the point of the value in adding a little term (I just grabbed Monday's numbers but sub-in yesterday if you'd rather the results will be too similar to matter either way). Assuming a static yield curve, consider the annual return of buying a 7 @ 1.69, receiving 4 semi payments of 8.45 and selling after two years for 101.5. This is roughly what's happening in an intermediate term fund. Next look if you'd done the same thing but bought the 5 @ 1.38, received 4 semis of 6.9 and sold after two years for 101.1. Finally, do the same with a 3 @ 1.01 4 semis of 5.05 and sold after two years for 100.4.

Buffetologist
Posts: 299
Joined: Sun Aug 22, 2010 8:58 am

Re: Critique Fixed Income Strategy - GNMA Dominance

Post by Buffetologist » Sat Apr 30, 2016 11:04 am

You know what? I've gone through the literature on Stable Value, and I've come to the conclusion that I simply don't have faith in viability of the life insurance industry in the face of low rates, Period. Just say no I say. All of these restrictions and rules and disclaimers yet I have no idea exactly what this thing holds. What's in an insurance contract? How do I know it's any good.

We are currently taking almost all of our risk in equities where I think we will get rewarded for it. I want that 35% devoted to protecting against the risks of equities. Credit risk is by and large equity risk also. This 65-35 split is the smallest we've ever had in equities. We were 100 equities until about 2010 when we started by Ally CDs in anticipation of college expenses. We've slowly ratcheted down from about 75-25 two years ago to 65-35 now. We ended up paying for college out of current income and the CDs became good deals as the rates dropped. My final Ally CDs mature in May.

I'm also realizing that even now, taking our after tax interest income and dividing it by the after tax value of our overall portfolio I get only 0.5% return from interest. If it was all cash in a 1% Ally savings bank using an after tax rate of 0.52%, I'd be getting 0.35*0.5 ~ 0.18%. So I'm fretting about 0.32% return on of my portfolio every year. That's less than many expense ratios. I don't think I'm being sufficiently compensated for additional credit risk or duration risk. The markets may think they compensate because there are so many entities -pensions and insurance companies - that simply have to take these risks. The whole mortgage bubble occurred because too many entities HAD to buy garbage and the investment banks and ratings agencies created the garbage especially for them. I simply do not have to get involved in this because I think can be happy with FDIC, NCUA and GNMA.

My mortgage dominates Spartan Intermediate Term Treasury
GNMA dominates my mortgage.
GNMA average maturity is 4.7 years, so even if rates go up, it doesn't extend duration by that much.

I think I'll just keep a large cash position to balance my GNMA position. If rates go up and stocks goes down, I have plenty of dry powder to rebalance into more equities. If intermediate term treasuries would pay at least as much as my mortgage, I might diversify then adding some intermediate term treasuries also. If another promotional CD appears that greatly exceeds the yield curve even after tax, I can jump. If rates go down, the equities will hopefully go up and protect me, and if they go below zero and equities crash, I'll take out a jumbo mortgage and let the mortgage company pay me, like that couple in the Netherlands is doing. LOL!

The 65-35 allocation is designed for a comfortable risk, as recommended by Bogle on Mutual Funds, and I think I'll leave my risk on the equity side.

Thanks for all of your opinions.

miles monroe
Posts: 1129
Joined: Mon Jan 20, 2014 12:14 pm

Re: Critique Fixed Income Strategy - GNMA Dominance

Post by miles monroe » Sat Apr 30, 2016 11:24 am

i'm certainly no fan of bob brinker -- really quite the opposite -- but during the years when he had a forum on his website on many occasions in response to questions he stated that he approved of a 50/50 allocation in retirement of total stock market index and gnma funds. just info.

User avatar
nedsaid
Posts: 8536
Joined: Fri Nov 23, 2012 12:33 pm

Re: Critique Fixed Income Strategy - GNMA Dominance

Post by nedsaid » Sun May 01, 2016 10:55 am

I own two GNMA funds, one at a workplace savings plan and another in a mutual fund IRA account. These have been good funds and have done exactly what I wanted. They sailed right through the 2008-2009 financial crisis. Every security has its unique set of risks, and I suppose that those risks have not shown up during my period of ownership. So I don't overdo it. My bond holdings are mostly plain vanilla Investment Grade Intermediate Term Bond funds, one of those funds is Vanguard Total Bond Market.
A fool and his money are good for business.

Valuethinker
Posts: 32925
Joined: Fri May 11, 2007 11:07 am

Re: Critique Fixed Income Strategy - GNMA Dominance

Post by Valuethinker » Sun May 01, 2016 11:35 am

In investing in US Mortgage Backed Securities (Agency securities) one has to have an understanding of prepayment and extension risk, negative convexity. Note Agency securities have no credit risk, whereas private label ones *do*. Prepayment risks are controlled for non US MBS.

Michael Lewis in Liar's Poker ain't a bad place to start. Ditto relevant chapters Larry Swedroe. I vaguely recall David Swensen has something to say about this.

On current plans FNMA and FMAC will no longer be able to issue MBS post 2018 (run out of capital). I don't think anyone knows what will happen to the mortgage/ MBS market, but we must presume credit risk (for non GNMA) will become a more significant issue (the fundamental terms of US mortgages would also change, presumably). As I say, no one knows-- the next Congress and President will jointly describe, and I don't know anyone who could reliably claim to understand the actions of *both*.

If you don't know what the above terms mean, or don't care, then I think you have to question whether you want to own significant amounts of these bonds ie a large proportion of your fixed income weighting. Because they could blindside you*.

VG GNMA is, I believe, an actively managed fund, so in a position to manage some of these risks.

A period of 35 years of falling interest rates, and a credit crisis based on housing price collapses, is not a good set of data to assess GNMA bonds against other types of bonds-- there's been a strong west wind almost the whole voyage (bar 1994, arguably).

* Vanguard Total Bond is 25% Mortgage Backed Securities. You could take Larry Swedroe's view (that given the complexities of those and corporate bonds (another 17% I think) you don't want to own this fund. Or the Taylor L. view which is that it's a good enough, and not worth trying to finesse it into say VG Intermediate Term Treasury. I probably lean towards the latter view, although I know that is incorrect for all the reasons discussed.
Last edited by Valuethinker on Tue May 03, 2016 10:33 am, edited 1 time in total.

Valuethinker
Posts: 32925
Joined: Fri May 11, 2007 11:07 am

Re: Critique Fixed Income Strategy - GNMA Dominance

Post by Valuethinker » Tue May 03, 2016 10:32 am

nedsaid wrote:I own two GNMA funds, one at a workplace savings plan and another in a mutual fund IRA account. These have been good funds and have done exactly what I wanted. They sailed right through the 2008-2009 financial crisis. Every security has its unique set of risks, and I suppose that those risks have not shown up during my period of ownership. So I don't overdo it. My bond holdings are mostly plain vanilla Investment Grade Intermediate Term Bond funds, one of those funds is Vanguard Total Bond Market.


All fine but note that 25% TBM is Mortgage Backed Securities (MBS). AFAIK no "private label" though (which are the ones with credit risk vs the US govt sponsored ones ie Agency bonds).

Valuethinker
Posts: 32925
Joined: Fri May 11, 2007 11:07 am

Re: Critique Fixed Income Strategy - GNMA Dominance

Post by Valuethinker » Tue May 03, 2016 10:42 am

https://books.google.co.uk/books?id=e-v ... BS&f=false

is Larry Swedroe's book chapter on MBS.

http://www.cbsnews.com/news/why-gnmas-s ... nd-choice/

It's not that these things are blow out dangerous (except under unusual circumstances ;-)) however they won't behave as an investor would like them to behave, under certain circumstances of interest rate changes. That lessens their value in a bond-equity portfolio w. rebalancing strategy.

User avatar
nedsaid
Posts: 8536
Joined: Fri Nov 23, 2012 12:33 pm

Re: Critique Fixed Income Strategy - GNMA Dominance

Post by nedsaid » Tue May 03, 2016 8:37 pm

Valuethinker wrote:
nedsaid wrote:I own two GNMA funds, one at a workplace savings plan and another in a mutual fund IRA account. These have been good funds and have done exactly what I wanted. They sailed right through the 2008-2009 financial crisis. Every security has its unique set of risks, and I suppose that those risks have not shown up during my period of ownership. So I don't overdo it. My bond holdings are mostly plain vanilla Investment Grade Intermediate Term Bond funds, one of those funds is Vanguard Total Bond Market.


All fine but note that 25% TBM is Mortgage Backed Securities (MBS). AFAIK no "private label" though (which are the ones with credit risk vs the US govt sponsored ones ie Agency bonds).


Indeed, Larry Swedroe dislikes Mortgage Backed Securities so much that he doesn't even like the Total Bond Index fund. He likes his clients to be in Treasury only funds the theory being that one should take risk on the stock side and not the bond side.

My own view is more sanguine, I owned GNMA funds, plain vanilla Investment Grade Intermediate Term bond funds, and Total Bond Market Index and they all just sailed through the 2008-2009 financial crisis. To my great surprise, TIPS got hit fairly hard during that time. Also, the government slapped guarantees on money market funds to stop what amounted to a run on the bank.

I have to smile in that I have seen a couple of threads on the horrors and dangers of GNMA funds but no one is starting similar threads about TIPS or Money Markets or Corporate Bond Funds for that matter. The fact that most classes of bonds got hit pretty good in 2008-2009 is no reason to not invest in them. The unique risks of those classes of bonds showed up in 2008-2009. Perhaps a market storm will show up that will show the unique risks of GNMA's, could happen but hasn't happened yet. That sports fans, is why you diversify.

Even if you love Ginnie Mae funds, don't go overboard.
A fool and his money are good for business.

User avatar
tennisplyr
Posts: 1305
Joined: Tue Jan 28, 2014 1:53 pm
Location: Sarasota, FL

Re: Critique Fixed Income Strategy - GNMA Dominance

Post by tennisplyr » Wed May 04, 2016 2:48 pm

I've had both Fidelity and Vanguard GNMA for a while and have no complaints.
Those who move forward with a happy spirit will find that things always work out.

VaR
Posts: 356
Joined: Sat Dec 05, 2015 11:27 pm

Re: Critique Fixed Income Strategy - GNMA Dominance

Post by VaR » Wed May 04, 2016 9:15 pm

I found Vanguard's paper, "The ABCs of MBS" to be worth a read:
https://personal.vanguard.com/pdf/s708.pdf

Executive summary. Mortgage-backed securities (MBS) constitute a large portion of the U.S. investment-grade taxable bond market. Although the performance of MBS is influenced by factors that affect all bonds—such as changes in interest rates—it is mainly the prepayment risk of MBS that causes them to perform differently from other bonds. This paper describes the relative characteristics of MBS, including their key risks and attributes, and the implications of investing in MBS alongside government and credit bonds in a taxable investment-grade bond portfolio. We also assess the drivers behind the past performance of MBS as well as the potential diversification properties of MBS going forward.

User avatar
Taylor Larimore
Advisory Board
Posts: 25981
Joined: Tue Feb 27, 2007 8:09 pm
Location: Miami FL

Beyond "Keep it simple."

Post by Taylor Larimore » Wed May 04, 2016 9:23 pm

Looking for some insightful critique beyond mantras like "Keep it simple" or that I'm "overthinking". I like overthinking.

Buffetologist:

Good luck, but you might hit my link below to see what experts say about "simplicity".

Best wishes.
Taylor
"Simplicity is the master key to financial success." -- Jack Bogle

Post Reply