Doc wrote:
Kevin, I just find the price/yield relationship in a Lehman type crisis to be just irrelevant.
This statement just baffles me--especially coming from someone who has seemed so sharp on "bond math" in other posts. The price/yield relationship for bonds is invariable--it always applies. You tell me the yield (and other relevant parameters, such as coupon, maturity and settlement), I can tell you the price. You tell me the price, along with the other relevant parameters, and I can tell you the yield. I'm sure you could do the same. End of story (at least it should be).
Doc wrote:In the three week period in late September through early November 2008 the S&P 500 was dropping at a rate of 1.5% a day. And you are telling me that a negative yield of 1% on a five year note is a significant factor.
If what you're asking is whether or not I think the 5-year Treasury yield could drop significantly below 0%--like -1%, -2% or further--then I guess my answer is that I don't know, but it's something we have not yet seen in this country, so it would be unprecedented. In all of my comments on this topic, my qualifier has been something like, "unless you think rates can fall significantly below 0% ...". So sure, to the extent that you think this is probable, continuing to hold intermediate-term or long-term Treasuries even as yields approach 0%, for a flight-to-safety hedge, could continue to make sense. I hope we don't see this occur, but hope is not a good basis for an investing strategy. And as Larry reminds us, we should not confuse the improbable with the impossible.
Doc wrote:Interest rates are driven by economic conditions with a little Fed tinkering. Bond prices adjust accordingly. In normal times interest rates are the independent variable.
Agree with the first two statements, but not necessarily the last. You can think of either price or yield as dependent or independent variable for a bond. They are just two ways of expressing bond value. I recall that I could enter a limit order to buy or sell a bond at Fidelity by specifying either price or yield, which is completely consistent with the way I think about bond price and yield now.
Having said that, I think the way people commonly think about it is as you describe. An example is the common way of expressing the duration rule of thumb as a percent change in price (dependent variable) given a one percentage point change in yield (independent variable). There are many other similar examples. But when it comes to the actual bond math, this is not the correct way to think about it. One starts to view it more the way it is after using enough PRICE and YIELD formulas in a spreadsheet.
Doc wrote:But the drop in the stock market had nothing to do with the economy. Treasury prices rose because of a flight to quality not a drop of 2% in the five year Treasury rate caused of some Zombie apocalypse that destroyed all the businesses in the country in a three week period.
Well, the first statement seems bizarre to me, but that's not directly relevant to this discussion, so I'll leave it alone. But just as you say Treasury prices rose because of a flight to quality, you could say with equal accuracy that Treasury yields fell because of a flight to quality. Again, just two ways to measure the same thing.
It is more common to think of increasing demand driving prices up, so just as it may be more common to think of yields as the independent variable when discussing things like duration, it might be more comfortable to think of price as the independent variable (in terms of price and yield) when discussing the effects of supply and demand. But from the pure math perspective, we can express the change using either price or yield without attributing either one as irrefutably the independent or dependent variable.
An imperfect analogy is that we can express mileage as either miles per gallon (common) or gallons per mile (uncommon). One goes up as the other goes down (so this is similar to price and yield), yet mathematically either one can be used to express the same thing. Just because it's common to express it one way does not make the other expression incorrect.
Doc wrote:The operative would here is flight. Irrational behavior not a duration x rate change equation.
But they are not mutually exclusive! Flight to safety is the underlying behavioral cause of yield and price changing, and duration just provides an approximation that defines the relationship between change in price and change in yield. The bond math applies exactly the same whether the underlying cause of the changes is flight to safety in a financial crisis or a more gradual change in the economy.
Doc wrote:Besides the Treasury yield recovered much faster than the stock market. The yield on the five was back at its mid-September starting point by June of the following year. The S&P didn't recover until some 15 months after that.
Sorry, don't get the relevance of this. Bond math applies to bonds in all cases.
Doc wrote:Kevin, I think we are in agreement that in a flight to quality situation Treasuries especially long ones, outperform CD's. The open question is if the cost associated with the better performance is worth it. The "insurance premium" if you like.
Yes, this certainly has been true in recent years. Looking at times longer ago, short-term Treasuries (3 month) sometimes did as well as or better than longer-term Treasuries (10 year) in big market downturns--at least if you look only at annual returns. I've shared this data before, using the Damodaran returns data as a source.
Since essentially all of the return for short-term Treasuries comes from interest--certainly over a one-year period--this is essentially saying that sometimes cash has outperformed the 10-year Treasury on an annual basis in big stock market drops. Since a good 5-year CD currently provides higher interest than cash, it follows that a good direct CD would have done even better than cash in these scenarios, unless you mark your CD to a mythical market value (since there is no secondary market for
direct CDs). If you object to the last part, just consider this an argument for holding some cash at 1% in an online savings account.
So we absolutely agree that based on recent history, Treasuries, especially longer-term, have provided a good hedge against falling stock prices, while at best CDs have not lost value--unless you had to do an early withdrawal to buy stocks, in which case they lost a little--on the order of 1%--still much less than the losses in corporate bonds during 2008.
But in the same breath that we say that, we also should say that CDs have consistently outperformed Treasuries if both have been held to maturity over at least the last 5-6 years, and that a good CD is guaranteed to significantly outperform a Treasury of the same maturity if both are purchased today and held to maturity. A good CD also is almost guaranteed to outperform a Treasury if rates increase enough to warrant doing an early withdrawal and paying the small penalty (the "almost" qualifier because of the remote chance that an early withdrawal will be disallowed). So that's two out of three in favor of the direct CD, leaving only the lack of potential upside to be capitalized on if selling before maturity after a decline in Treasury yields (e.g., to buy stocks). I guess your way of acknowledging this is to reference the insurance premium you're paying for Treasuries.
Doc wrote:We also agree/ that you don't have to have all your FI portfolio in the single best sector for an equity crash.
Absolutely. I'd go further and say that you don't have to have
any of your FI in this asset class (Treasuries) if you're willing to give up the potential bump in a flight-to-safety scenario, which is the case for me.
Doc wrote:It seems to me that what we should be addressing is if the rest of the portfolio should be CD's or Corporates.
Of course this question doesn't apply to me, since I hold CDs and some corporates but no Treasuries (other than what's in my investment-grade bond funds).
I recall from a recent post that Larry said that his firm uses CDs almost exclusively for the tax-advantaged portion of their clients' portfolios, since the yields are so much better than Treasuries. They even use CDs in taxable accounts of wealthy clients since after-tax yields can be even higher than muni yields, especially for shorter maturities. So
I'm wondering where Larry comes down on the use of Treasuries as a hedge against financial crisis--from my understanding of what he's said, it seems that he might be in the same camp that I'm in, and simply is willing to give up the potential price bump of Treasuries in return for the much higher guaranteed yields of CDs.
Kevin
If I make a calculation error, #Cruncher probably will let me know.