Rising rates: "What - Me Worry?"

[BTravlin]

IMO it's pretty silly to argue that one shouldn't rebalance because one of the asset classes you might buy is "too rich". That basically disavows the whole idea of maintaining an AA. Yet I see these comments all the time. The fact is, we don't know which asset class will do well over then next year. I don't believe in "anticipatory" not rebalancing.

I rebalanced. I bought some bonds. My equities went up much more than my bonds in 2012, so I ended up buying some bonds. If my bond funds hadn't gone up so much as well, I would have bought a lot more. Because they did go up, I bought less. When they (eventually) go down, I'll buy more - assuming my equities don't go down as much. That's how rebalancing works.

I understand what you're saying and agree. Many of the so called investment pros in this article say equities have gone up so they should rebalance with bonds but believe bonds are too risky right now due to distortions caused by QE.

I say if that's the case then equities have also been distorted by QE and their argument is therefore moot.

 

[Lsbcal]

I actually rebalanced today and that equity money will go into a short term bond fund, eventually to be used for spending. That's because I make sure my equities are at the max tolerable allocation so I don't want any more and don't have to worry that I might be missing any action. My focus is on the bigger slice of equities, not on the little bit that is "over my line".

But the real question as discussed in the OP is how your bigger slice of intermediate bonds might perform in a rising rate environment over the future months and years. And even that bigger slice should really be viewed in the context of the overall portfolio performance, as several posters have mentioned here.

Rebalancing is usually only a tiny fraction of one's portfolio and doesn't contribute much to the overall results. That's true for me anyway.

My 2 cents on rebalancing.

 

[brewer12345]

This is not a hugely stressful series of tests (50BP changes in the shape of the curve and 100BP upward shift), but it is striking to see how large the value changes are for such small rate movements: Interest rate risk in the bond market — FactSet Research Systems

 

[pb4uski]

There was an article in the WSJ the other day stating many market pros are having second thoughts on rebalancing this year. They say the recent run up in equities points to the need for rebalancing but many consider bonds too risky right now to carry through with the usual strategy and will therefore stay heavy to equities.

While on the one hand I understand their reasoning, on the other hand it just seems like another version of timing. Thoughts anyone?

I recall a couple different times over the last 20 years where I didn't rebalance out of fear (the last was buying stocks in 2008) and in both cases wish I had, so I have become a believer in rebalancing no matter what. That said, I recently rebalanced and am still pretty close to my target AA so no actions at this time.

 

[Lsbcal]

This is not a hugely stressful series of tests (50BP changes in the shape of the curve and 100BP upward shift), but it is striking to see how large the value changes are for such small rate movements: Interest rate risk in the bond market — FactSet Research Systems

I'm not sure how to go from that article to how a rate rise would affect a fund based on Barclays U.S. Aggregate Bond Index as I understand Vanguard's Total Bond Index VBTLX is based on. The original table I showed in the OP has VBTLX in it with past rate rise performance.

Any ideas how to relate such a stress test in that article to an intermediate bond fund like VBTLX?

 

[audreyh1]

I understand what you're saying and agree. Many of the so called investment pros in this article say equities have gone up so they should rebalance with bonds but believe bonds are too risky right now due to distortions caused by QE.

I say if that's the case then equities have also been distorted by QE and their argument is therefore moot.

Yep - exactly!

 

[nun]

+1, another good read IMO, thanks.

If you expect the price of your bonds to go up then the good times are probably at an end, but as the vast majority of return from bonds comes from the income they generate if you are not going to be selling any time soon just hold on and any reduction in price due to increasing interest rates will eventually be compensated by increased income.

 

[brewer12345]

I'm not sure how to go from that article to how a rate rise would affect a fund based on Barclays U.S. Aggregate Bond Index as I understand Vanguard's Total Bond Index VBTLX is based on. The original table I showed in the OP has VBTLX in it with past rate rise performance.

Any ideas how to relate such a stress test in that article to an intermediate bond fund like VBTLX?

Um, did you read the linked article? It discusses changes in the securities which compose the Barclays Aggregate (they have gotten longer in maturity, especially in the corporate sector - danger Will Robinson) and a 100BP increase in rates (they call it a 100BP Bear Parallel) would produce a loss of almost 5%. 100BP move does not look that stressful to me, so we can extrapolate that a real hit of 200 to 300BP rate increase would be quite painful.

 

[Lsbcal]

Um, did you read the linked article? It discusses changes in the securities which compose the Barclays Aggregate (they have gotten longer in maturity, especially in the corporate sector - danger Will Robinson) and a 100BP increase in rates (they call it a 100BP Bear Parallel) would produce a loss of almost 5%. 100BP move does not look that stressful to me, so we can extrapolate that a real hit of 200 to 300BP rate increase would be quite painful.

I'm not questioning your thinking or the article info. Also I never set myself up as an expert in this stuff, am always willing to learn new things. I wasn't quite sure what the Bear Parallel referred to, thanks. How the 100BP Bear Parallel relates to a 5 year Treasury constant maturity increase I'm still unclear on. And maybe that correlation is not important anyway.

But wouldn't we do just as well by taking the average duration for VBTLX and multiplying by the rate increase? For a 100BP change that would give us 5.2 x -1.0% = -5.2% rate rise loss. If that rate rise were for 1 year then we get to add back in the distribution income which might be something like 2%. So the net for the year might be -3.2%.

Here is some very recent and relevant data for the last month. The 5 year Treasury constant maturity went up roughly from 0.72% to 0.90% (18BP) during this time.

The 1 month returns for a selection of fund using M* data:
-0.57% IEI, iShares 3-7 year Treasury
-0.87% VBTLX, Vanguard Total Bond Mkt (uses the Barclays Agg index)
-0.17% PTTRX, Pimco Total Return
-0.10% BOND, Pimco Total Return ETF
-0.14% DODIX, Dodge and Cox Income

So from this data we might roughly infer that a 100BP rise in the 5 year Treasury would give us a loss of -4.8% in VBTLX. Sounds somewhat consistent with the previous estimates. Also the 18BP change last month implies a change in VBTLX of 5.2 x -0.18 = -.94% which is near the actual number (probably need to add a little yield back for the 1 month distribution yield).

I'm just glad to see that the active funds PTTRX, BOND, and DODIX did not do as bad as either the Treasury fund or the Total Bond Mkt fund in the last month. Maybe in a stiffer rate rise they will do better then the Barclays Agg Index. Who knows.

 

[samclem]

IMO it's pretty silly to argue that one shouldn't rebalance because one of the asset classes you might buy is "too rich". That basically disavows the whole idea of maintaining an AA.

Yes, I see the point. But the underlying logic behind allocations and rebalancing assumes that market forces are at work and that the various asset classes are fairly priced (given what the market "knows" about anticipated future values of each). I don't think that's the case with bonds right now--the Fed has its thumb on the scale and has been pressing down to keep interest rates low. When it lets up the market will fairly price them: bond yields will go up, and bond prices will fall (esp for longer terms).

I rebalance faithfully and don't try to time the stock market. But, I don't think it's inappropriate to reconsider our allocation %ages if we believe the rules have been altered, and it's clear that's what we're experiencing now.

I intend to reset to a more "normal" bond allocation once the Fed is done with its (openly acknowledged) QE interest rate manipulation. I suppose if the government/Fed had declared a war on dividends and stock appreciation and put in place policies to limit them, I would have reduced my equity exposure and put more into bonds.

 

[brewer12345]

How the 100BP Bear Parallel relates to a 5 year Treasury constant maturity increase I'm still unclear on. And maybe that correlation is not important anyway.

The bond index is not composed of 5 year treasuries or the 5 year treasury CMT index. It is a pile of all kinds of bonds, many of which have maturities way beyond 5 years (I have been seeing 60 year bonds in the investment grade new-issue market) and a large chunk (~30%) of which are mortgage bonds where a quoted duration is just somebody's guess based on expectations about mortgage borrower behavior. In addition to a uniform yield curve shift in one direction or another, values can be substantially affected by changes in the shape of the yield curve. Rates out to 10 years could stay about where they are while the 10 to 30 year part of the curve rises a lot and the index would take a lot of pain.

 

[Lsbcal]

The bond index is not composed of 5 year treasuries or the 5 year treasury CMT index. It is a pile of all kinds of bonds, many of which have maturities way beyond 5 years (I have been seeing 60 year bonds in the investment grade new-issue market) and a large chunk (~30%) of which are mortgage bonds where a quoted duration is just somebody's guess based on expectations about mortgage borrower behavior. In addition to a uniform yield curve shift in one direction or another, values can be substantially affected by changes in the shape of the yield curve. Rates out to 10 years could stay about where they are while the 10 to 30 year part of the curve rises a lot and the index would take a lot of pain.

It is an interesting point that the Barclays Agg Index has changed over recent years. M* did an article some months back that discussed how this index has drifted towards a higher percentage of Treasuries. I keep an eye on the index but don't own Total Bond Market (VBTLX). I wonder how VBTLX's average duration has changed and whether that is a reasonable way to judge the index's rate sensitivity.

 

[AndrewJackson]

Hmmm...

The problem with these comparisons to the past is that in most of the past cases rates were nowhere near the current extremely low rate environment. We are a bunch of standard deviations out on the curve. As a result, I choose to be more careful regardless of what the past sggests.

Spot on.

 

[audreyh1]

Yes, I see the point. But the underlying logic behind allocations and rebalancing assumes that market forces are at work and that the various asset classes are fairly priced (given what the market "knows" about anticipated future values of each). I don't think that's the case with bonds right now--the Fed has its thumb on the scale and has been pressing down to keep interest rates low. When it lets up the market will fairly price them: bond yields will go up, and bond prices will fall (esp for longer terms).

I rebalance faithfully and don't try to time the stock market. But, I don't think it's inappropriate to reconsider our allocation %ages if we believe the rules have been altered, and it's clear that's what we're experiencing now.

I intend to reset to a more "normal" bond allocation once the Fed is done with its (openly acknowledged) QE interest rate manipulation. I suppose if the government/Fed had declared a war on dividends and stock appreciation and put in place policies to limit them, I would have reduced my equity exposure and put more into bonds.

I disagree that AA and rebalancing assumes that asset classes are ever fairly priced. IMO it assumes the opposite - that asset classes are often mispriced, high or low, but that the pendulum eventually swings the other way. Now normally we might not have most asset classes overpriced as seems to be the case today with all the QE, and or most asset classes underpriced as appears to have occurred in fall of 2008. But even in these cases, not all asset classes are overpriced or underpriced equally, so diversification still seems to help.

As I have mentioned before, I feel better about the current situation because I've maintained my AA for a very long time so a lot of my holdings were bought long ago. FWIW long-term bonds were never part of my AA.