In Part 1 of this series, I illustrated why TIPS provides some deflation protection, though not as good as nominal bonds, and why both did terribly at the height of the '08 crisis. At this point it's worth reiterating the old wisdom: bonds are good for bad times and go opposite of stocks. Except, of course, when it's not true. And it is when the old wisdom is not true that one needs to pay special attention because the market is trying to tell you something different and interesting. But let's finish the TIPS-nominal analysis during normal times before going into the more interesting part.
4. TIPS offers only half inflation protection.
When inflation heats up, both interest payments and the expected principal payback for TIPS go up, hence "inflation protection" which makes it better than nominals in this case. But the "real interest rate" usually goes up when inflation heats up, since demand for capital usually increases. TIPS protect you against the CPI part, but not the "real interest rate" part. This point has been explained by many in the media and blogosphere, unlike the deflationary protection which has attracted much less attention, so I'll not dwell on it further.
5. Holding bonds to maturity does not give you any more certainty vs. selling it on the secondary market.
A common falsehood in the financial advisory universe is that if you hold a bond to maturity, you know how much you'll get back (assuming no default), as opposed to selling it on the secondary market where you may have to take capital loss. Again, funny if not so sad.
If a bond sells on the secondary market at a discount, there's a reason for it -- people want cash now as opposed to later. Why people want cash now as opposed to later could be either inflation protection or panic/crisis. In this case, does your "known" amount of interest payment and principal payback provide any extra value? Absolutely not. If you bet the market is wrong, then you're making a directional bet on the future of your current investment, which is no different from making a new directional bet.
In more concrete terms, if a bond is selling at $80 now, it's only because either the market expects all its future interest and principal payback to be worth $80 today or people would rather have $80 cash today as opposed to $110 in a few years. If you think the market is wrong, which you're most certainly entitled to and could very well be right, then you may decide to hold it rather than selling. But, for Mt. Everest's sake, don't hold it because holding to maturity protects your investment -- it doesn't.
6. Buying bonds in the primary market does not give you any advantage.
Another financial advisory falsehood that gets me worked up is that buying bonds at the secondary market is a bad idea since you may have to pay premium. I mean, do people really think the markets are so stupid that this could possibly be true?
7. But do watch out for spreads and commission, and of course, tax.
Above being said, it is very true that, because of the general lack of liquidity in secondary bond markets, the bid/ask spread can be outrageously wide for individual investors. Before you buy bonds, even if from the primary market, you should think hard what the chances are you'll hold it to maturity. If you sell it on the secondary market, the bid/ask spread (and usually to a much lesser degree, commission) could easily take a huge bite out of all your gains, and quite possibly more. So, 2-5% for even big corporate names is common. Treasuries/TIPS are better, but given today's infinitesimal coupon/yield, a 1% spread is quite painful to swallow.
It is in this sense that buying/selling bonds on the secondary market should be thought out carefully. For TIPS/treasuries, Uncle Sam tries to make it easy for you to buy but not to sell -- like all other crooks, no surprises there. And if you do end up in the secondary market, don't buy at the offer or sell at the bid unless you're talking odd lots (less than $100k worth of original principal); at least try some price in the middle first. You might be surprised.
Another factor is tax. Assume you could find two roughly equivalent bonds, one in the primary market and another in the secondary. The difference is that the one in secondary shifts some capital gain/loss into interest income or the other way around. It's similar for selling in the secondary markets vs. holding it to maturity. For anything less than $1M, I think the difference is unlikely to be worth the accountant fee or the brain cells dying of exertion from all the thinking. But you be the judge for your own money.
8. The market shows that the sweet spot for TIPS and nominals is CPI 1-1.5.
That was all theory and logic. How does the market think of it?
When talking about inflation, it's very important to separate "real interest rate" from inflation. The following chart is the 5-year CMT yield since Jan 2003 vs the "real interest rate," which for our purpose here is simply the TIPS yield. Don't be too impressed by how tight and straight the blue line is; it is so by definition. The point is that it shows, as mentioned earlier, nominals are also riskier than TIPS with respect to "real interest rate" moves. The stray group of orange dots are from the Sept-Nov 2008 period, which shows how out of whack and "nonsensical" it was.
How do TIPS and nominals perform with respect to inflation? The following chart is the TIPS and nominals yields vs. the 5-year CPI expectation (annual %), which is CMT-TIPS.
The dots left of CPI=1.0 are again from the crisis period. If you ignore them, then it shows that TIPS are less risky than nominals, again consistent with theory. But if you do include them, even if ignoring the far-fringe points, you'd be led to some interesting observations:
- There's a sweet spot for both TIPS and nominals in terms of CPI, which is at about 1.5, or more vaguely between 1 and 1.5. If CPI falls below this, it's no longer "the lower the interest rate, the better for bonds;" instead, the market jumps into crisis mode, be it credit risk or liquidity crunch. This is a new insight as far as I know.
- Both TIPS and nominals have negative convexity, in terms of price, with respect to CPI.
- While TIPS are less risky than nominals in the "normal" environment, in or above the sweet spot, it is more risky below the sweet spot while nominals offer excellent protection (the line is almost flat). Is this only because the change in inflation expectation in late 2008 was so abrupt and severe? Is this an arb or a trap?
In summary, TIPS are much more complicated than their name suggests. It's sad that the Treasury designed something so misleading and complicated. But still, it's one of the best things we have to measure and hedge against inflation. You just need to look beyond the name.
Disclosure: I am long GLD, SLV, DBA.