Best way to buy long term treasuries

[greg9840]

Hello everyone,

Based on everything I have read about the permanent portfolio, it seems close to unanimous that people should buy actual long term treasuries, when possible, as opposed to buying a long term treasuries fund.  But I am seeing conflicting information on how to buy treasuries.  Some people say that when you start your PP, you should buy all 30 year treasuries, and 10 years later, when they become 20 year treasuries, sell them and buy 30 year treasuries again.  But I think we should probably not ignore the fact there is a small, but significant, difference between how 30 year treasuries and 20 year treasuries behave.  I don't like the idea of going through, for example, an economic climate of extreme interest rate changes, and arbitrarily holding all 30 year bonds or all 20 year bonds.  Does it make more sense to buy a ladder of long term treasuries where you buy 10% 30 year treasuries, 10% 29 year, 10% 28 year, all the way down to 21 year treasuries.  Then, you would have an average of around 25 year treasuries at all times and each year, you sell the bonds that hit the 20 year treasury mark and use the money to buy 30 year treasuries.  Does anyone have any other ideas?  Has anyone back tested different long term treasury strategies?

Thanks!
Greg

[dualstow]

Some people say that when you start your PP, you should buy all 30 year treasuries, and 10 years later, when they become 20 year treasuries, sell them and buy 30 year treasuries again.

Hi Greg.
It's not just that some people say this. It's canonical. Those were Harry Browne's instructions in his books. And, I think the source of dissent has less to do w/ 20-YR vs 30-YR instruments and more to do with the pain of selling treasuries at a loss. Everything else gets sold high or bought low, but bonds may have to be sold simply because their time is up, and that feels as counter-intuitive as starting a pp feels when stocks are doing so well.

[Ad Orientem]

I think it best to avoid dogmatism with the PP. There are some areas where I have really strong convictions (like not substituting individual stocks for the equity part), but there are many people who index their LTTs including yours truly. And no I don't think it is heresy.  Further, I am pretty sure that Harry endorsed SHY on one of his programs as an acceptable substitute for cash. It depends on your risk tolerances and what will keep you awake at night. I am prepared to accept a very slight level of added risk for the convenience of not having to deal with details. Others are not.

Your mileage will vary.

[smurff]

Somebody could make a Harry Browne Permanent Portfolio Bible, with transcripts from his shows where he talked about the PP and answered questions about whether this or that investment would be appropriate, any speeches he made about the PP, and excerpts from his books about the PP.  Then we could quote him, chapter and verse.  Sort of like Nostradamus.

Of course, with all the financial product innovation and the wisdom from this forum, it would eventually be out of date as a practical guide, but for the dogmatists it would be helpful to have.

[Tortoise]

Somebody could make a Harry Browne Permanent Portfolio Bible. . .

Would I be able to read it on my computer screen or even print it out and read it in my easy chair?

[smurff]

Ideally, yes, but be sure to print it out on bible paper--otherwise the pages assembled together would be like an unabridged dictionary and be too heavy for leisurely reading in an easy chair. 

But it would be perfect for an e-book, with cross-referencing, the ability to look up words and phrases ("fraud," "investment vs. speculation," "money you cannot afford to lose," etc.).  A  paper bible (p-bible)  would need its own concordance, just to be able to find stuff.

[dualstow]

I think it best to avoid dogmatism with the PP. There are some areas where I have really strong convictions (like not substituting individual stocks for the equity part), but there are many people who index their LTTs including yours truly. And no I don't think it is heresy.  Further, I am pretty sure that Harry endorsed SHY on one of his programs as an acceptable substitute for cash. It depends on your risk tolerances and what will keep you awake at night. I am prepared to accept a very slight level of added risk for the convenience of not having to deal with details. Others are not.

Your mileage will vary.

I agree completely. I only meant what I wrote in a specific context: making it clear that this is not a case of "some people say you should hold 30-yr bonds directly."

While I don't remember Browne endorsing SHY, he was clearly very excited when a listener told him about TLT, and he was anxious to learn more about it. then again, the government was not even issuing 30-year bonds when he made that discovery.

Of course, Harry's dead and we are left to ourselves to decide when and if things have changed to a point where we cannot blindly stick to the original prescription. (That's why this forum and the new book are so valuable). I don't mean changes in economic "weather" so much as changes in things like what's going on under the hood of TLT.

For what it's worth, I hold both TLT and pure LTTs.

[greg9840]

I think Harry Browne thought funds such as TLT make it easier for people to use the PP strategy.  But he was always very cautious and focused a lot on all of the risks.  Funds like TLT add another layer to the risks in the event iShares makes a mistake, commits fraud, or fails completely.  There are 2 totally separate set of risks: you have the day-to-day investment risk of asset values going up and down, and you also have the risk of a catastrophe, which is obviously much less likely to happen, but it is likely enough, at least in the eyes of Harry Browne, for him to strongly recommend using as few layers as possible, splitting holdings between multiple brokers, and storing assets/gold in another country, just in case things get crazy in your home country.  Many people choose to only follow the investment strategy part, but ignore the risk of a catastrophe part.  Does anybody know if the people who lost money with MF Global did not get 100% of their US treasuries back?

[greg9840]

Does anybody on this forum ladder their long term treasuries?  I know HB said to buy 30 year treasuries and sell them 10 years later, but doesn't it make sense to build a 10 year ladder with average maturity of 25 years?

Or, does it make sense to buy only 30 year treasuries and sell them all after 5 years instead of waiting 10 years?  I am just trying to figure out of there was logic behind HB saying to sell them in 10 years or if it was maybe a bit arbitrary.  I would guess that selling 30 year treasuries after 5 years instead of 10 years would generate higher returns.

Thanks,
Greg

[dualstow]

...
  I am just trying to figure out of there was logic behind HB saying to sell them in 10 years or if it was maybe a bit arbitrary.  I would guess that selling 30 year treasuries after 5 years instead of 10 years would generate higher returns.

I think it was a balance between (a) having something long enough to be sufficiently volatile to protect against stock losses and (b) keeping trading frequency low. So, (a) under 20 years is out and (b) only trading every ten years (not counting rebalancing) means less taxes or less frequent selling at a loss and less trading to do.

[Tortoise]

Does anybody on this forum ladder their long term treasuries?

Yes. Since I'm still in the accumulation phase, I just buy a new 30-year bond whenever I need to add to the bond portion of my PP. This is slowly creating a ladder of sorts. Setting up a ladder from the beginning would have been another acceptable approach, but I just decided to do it this way because it simplified getting started.

I know HB said to buy 30 year treasuries and sell them 10 years later, but doesn't it make sense to build a 10 year ladder with average maturity of 25 years?

Or, does it make sense to buy only 30 year treasuries and sell them all after 5 years instead of waiting 10 years?  I am just trying to figure out of there was logic behind HB saying to sell them in 10 years or if it was maybe a bit arbitrary.  I would guess that selling 30 year treasuries after 5 years instead of 10 years would generate higher returns.

To my knowledge, there was nothing magic or quantitative about HB's recommended 20-year maturity threshold for selling long-term treasuries. Simplicity was probably his main consideration. He realized that most people aren't investing geeks who enjoy setting up bond ladders, so he probably wanted them to read his book and think to themselves, "After I buy the long-term bond, it'll be 10 whole years before I even have to touch it again? That's pretty darn simple. Even I can do that!"

If you roll your bonds over after 5 years instead of 10 years, you're talking about an average time to maturity of 27.5 years instead of 25 years. That's not a big difference. The yield and duration of 27.5- and 25-year bonds are very similar. So in terms of volatility and return, either approach for laddering is fine.

In a taxable portfolio, however, the two laddering approaches are potentially different in terms of taxes (as dualstow briefly mentioned). Selling after 5 years requires selling one-fifth of your total bond allocation every year, whereas selling after 10 years requires selling only one-tenth of it every year. If you want to avoid getting hit with big capital gains, selling the bonds after 10 years tends to lessen the blow. (In a tax-sheltered account like an IRA or 401(k), this isn't an issue.)

[greg9840]

Thanks for all the great feedback.  If my 30 year treasuries have dropped in value after, say, 5 years, (when there are 25 years remaining to maturity) might it make sense to sell the treasuries at that time to lock in a tax loss (this is in a taxable account), and then use the proceeds to buy new 30 year treasuries?  Maybe it would make the most sense to look at the values of the bonds in 5 years and, if they are up in value, just keep them, but if they have dropped from the purchase price, then sell them.  Any thoughts?

Thanks!

[barrett]

Just came upon this old thread and I wanted to ask for advice. I just started a PP this year and grabbed all 11/15/43 bonds (3.75% coupon) for the 25% LTT allocation. Any problem with doing that? I mean, do I need to ladder them or just sit on my backside for the next ten years? These are all in tax-deferred accounts. Thanks.

[Ad Orientem]

Just came upon this old thread and I wanted to ask for advice. I just started a PP this year and grabbed all 11/15/43 bonds (3.75% coupon) for the 25% LTT allocation. Any problem with doing that? I mean, do I need to ladder them or just sit on my backside for the next ten years? These are all in tax-deferred accounts. Thanks.

Nothing wrong with that approach. Purely for convenience, I prefer ETFs for everything but gold. But yours is the way to go for those seeking that extra degree of safety.

[barrett]

Thanks Ad Orientem. Because it's the asset I understand the least, I need to be talked down from a ledge every few days on the LTT's. Appreciate the help.

[AdamA]

Just came upon this old thread and I wanted to ask for advice. I just started a PP this year and grabbed all 11/15/43 bonds (3.75% coupon) for the 25% LTT allocation. Any problem with doing that? I mean, do I need to ladder them or just sit on my backside for the next ten years? These are all in tax-deferred accounts. Thanks.

Just sit on your backside.  Usually the best approach to investing, in general.

[Fragile Bill]

If my 30 year treasuries have dropped in value after, say, 5 years, (when there are 25 years remaining to maturity) might it make sense to sell the treasuries at that time to lock in a tax loss (this is in a taxable account), and then use the proceeds to buy new 30 year treasuries?  Maybe it would make the most sense to look at the values of the bonds in 5 years and, if they are up in value, just keep them, but if they have dropped from the purchase price, then sell them.  Any thoughts?

If the bonds are dropping in value, then interest rates are going up.  Why wait five years to look at the bonds in a rising interest rate environment?  In such an environment, wouldn't it make sense to replace them annually to earn as much interest income as possible to offset loss in the value of the bonds?

[Ad Orientem]

If my 30 year treasuries have dropped in value after, say, 5 years, (when there are 25 years remaining to maturity) might it make sense to sell the treasuries at that time to lock in a tax loss (this is in a taxable account), and then use the proceeds to buy new 30 year treasuries?  Maybe it would make the most sense to look at the values of the bonds in 5 years and, if they are up in value, just keep them, but if they have dropped from the purchase price, then sell them.  Any thoughts?

If the bonds are dropping in value, then interest rates are going up.  Why wait five years to look at the bonds in a rising interest rate environment?  In such an environment, wouldn't it make sense to replace them annually to earn as much interest income as possible to offset loss in the value of the bonds?

You own LTTs for their volatility, not for their yield. You sell the bonds when they rise in value to the point where they constitute 35+% of the aggregate value of your portfolio. If you want to do some quick tax loss harvesting that's one thing. But there is risk in that. You could end up buying the bonds back at a higher price.

[Fragile Bill]

I understand the concept of re-balancing.  I'm referring to the notion of replacing, in a rising interest rate environment, year old bonds paying a lower interest rate with new bonds paying a higher rate.  For example, if interest rates increase steadily to 8% in late 2017, I'm calculating a 39% loss in the bond portion of the PP, and that's when I replace my bonds annually.  If I hold onto the same old 3.60% bonds until late 2017, I'm showing a 47% loss in the bond portion of the PP.

[Pointedstick]

I understand the concept of re-balancing.  I'm referring to the notion of replacing, in a rising interest rate environment, year old bonds paying a lower interest rate with new bonds paying a higher rate.  For example, if interest rates increase steadily to 8% in late 2017, I'm calculating a 39% loss in the bond portion of the PP, and that's when I replace my bonds annually.  If I hold onto the same old 3.60% bonds until late 2017, I'm showing a 47% loss in the bond portion of the PP.

The efficiency of the bond market makes it a wash at best. You'd take a loss on the old lower-interest bonds you sell, and pick up new, higher interest bonds for more money, with the the higher interest payments covering your loss depending on the holding period. With a short enough holding period, you're just continuously buying high and selling low.

If interest rates are 8%, think of how insane your cash is going to be. Stock dividends are going to be pretty high too. But 8% interest in 2017 sounds far-fetched at best to me.

[Fragile Bill]

Still not sure that I follow.  The performance of the bond portion of the PP, in a rising interest rate environment, equals the decline in the present value of the bonds, partially offset by interest earned.  If interest rates go from 3.60% today to 4.00% in a year, then let's say my $25,000 bonds have a PV of $23,000 a year from now.  I can either keep my $23,000 bonds for another year and earn 3.60%, or sell them and buy new $23,000 bonds that earn 4.00%.  Tell me again why that wouldn't maximize the performance of the bond portion of my PP...

[Fragile Bill]

I think I just answered my own question.  The year old bonds may now be worth only $23,000, but the 3.60% interest is being paid on $25,000.  The new $23,000 bonds will be paying 4.00%,  but 4.00% on only $23,000.  That must be the "wash" you are referring to.