How much would you pay to lend money to the government? Most of us have this arrogant idea that the government should pay us to borrow our money. And yet, last week a Treasury auction of $10 billion in 5-year bonds resulted in a price that will yield negative 0.55% to their new owners.
It is not quite as crazy as it sounds. These are Treasury Inflation Protected Securities (or TIPS) that will yield inflation plus some stated interest rate. So these bonds are set to return to their owners inflation minus 0.55% over five years. Given that normal unprotected five year bonds are currently paying only 1.18%, this implies a five year average inflation rate of 1.73%.
Annual inflation over the past five years has averaged 1.83% and over the past twenty five it has been 2.82%. If you think inflation over the next five years will be higher than 1.73%, then the TIPS, negative interest and all, are a better bet than the regular Treasurys.
So it is not crazy after all.
Or is it? Locking in the rate of inflation as an investment return ought not to be much of a challenge. All you need to do is create a portfolio of real stuff, that is, things that are neither dollars nor denominated in dollars. Commodities, real estate, foreign currencies and/or bonds, collectibles, and even equities count. The things you buy do not even need to produce any return, i.e. make any income. All you ask is that they retain real value.
At the risk of giving away a brilliant idea, I am wondering why there isn’t an inflation ETF available. (You can get pretty close with a few of them, though.) If a person can get inflation as a return, why would anybody settle for inflation minus 0.55%? Indeed, owning actual things for five years has considerable tax advantages over getting bond income.
There turns out to be a good answer to that question, but before I unveil it permit me to digress a little to point out how unhelpful the business media was when reporting and discussing the negative yield auction result. Useless financial commentary is the main topic of this blog, after all.
Many outlets reported the negative yields as, of all things, a vote of confidence in the Fed. Bloomberg told us that it was a “bet the Federal Reserve will be successful in sparking inflation.” The Washington Post echoed that with only a hint of uncertainty, sneaking a question mark into their headline Sign of faith in Fed? Yield on inflation-protected Treasurys dips below zero. The Wall Street Journal was more obviously ambivalent, subheading its story “Big Demand for Bonds Suggests Fed Is Winning Deflation Battle” but then in the first sentence citing “growing fears of rising prices.”
For those of us who have been conditioned over many decades to think of the Fed as an institution that exists primarily to fight inflation, the idea that they are now purposefully trying to cause it, and indeed might not succeed, takes some getting used to. It is like reading that observers are now confident that the fire department will be able to burn down a few houses after all.
Then there was the passel of reports and blog posts seeking to calm those unnerved by this development, as if it were a headline grabbing disaster. Larry Swedroe gave us Why the Concern over Negative TIPS Yields Is Overblown, which seemed to imply more anxiety than was immediately visible.
Swedroe makes the not unreasonable point that unless you expect particularly low inflation in the coming years, the real rate on Treasurys has been pretty obviously negative for a while now. He even argues that, everything else equal, TIPS are probably a better bet than regular Treasurys because he expects inflation to be higher than the implied 1.73% prediction.
Okay, but he ignores what I consider to be a relatively obvious objection to locking in inflation minus 0.55% as a return, that you can relatively easily lock in inflation minus nothing instead.
An article in the Wall Street Journal a few days after the auction almost, but not quite, raised this objection. Mostly, it demonstrated how mysterious TIPS are to WSJ reporters.
Investors might want to consider other options for inflation protection.
That is because Treasury Inflation-Protected Securities, or TIPS, tend to have a low "correlation" to the consumer-price index, the main inflation gauge. The correlation between the two from 2002 to 2009, for example, was just 0.21%. A correlation of 1 means two assets move in perfect lockstep, a correlation of -1 means they move in opposite directions, and a figure of zero means they are uncorrelated.
Why the quotes around correlation? Is it meant ironically? Without running the numbers myself, I am guessing that the correlation between the CPI and the secondary market prices for TIPS is 0.21. (Not 0.21%.) And that should not be surprising or worrisome. Although the face value of TIPS goes up with inflation, the market price of the bond, like any bond, goes up and down with changing interest rates. Further, TIPS will go up and down as inflation/deflation expectations for the future wax and wane.
More to the point, if you buy a TIPS for the $1000 face value when issued and hold it to maturity, and if the CPI has been up 10% in the meantime, you will get $1100. Guaranteed.
Even better, TIPS have an odd feature that is often overlooked. No matter what happens to the CPI, the face value of the bond, what the government will pay to redeem it when it matures, will never be less than the $1000 issue price. So if the CPI goes up 10%, you get $1100, but if it goes down 10%, you still get your $1000 back, not $900. Thus, the inflation protection in TIPS is of the form of a call option on the CPI. You get paid if prices go up, but do not suffer if they go down.
And this is why TIPS are, arguably, much better than owning a portfolio of real things whose value tracks the CPI. Should you be wrong in your prediction of inflation, and deflation rears its ugly head, the real things will decline in price. Not so with the TIPS. Indeed, the $1000 the government will return to you will be more valuable than the $1000 you gave it years before.
Does this mean that TIPS with a yield of –0.55% are a good deal? Is 55bps a good price to pay for this one-sided inflation protection?
Probably not. Personally, I think the likelihood of deflation over a five year period is close to nil. 1970s style inflation is also unlikely to materialize, but deflation is, for all the buzz, a rare beast on these shores. The last time the CPI was down for a single calendar year was 1954. (2008 was +0.1%.) The CPI has not been down on a five year basis since we abandoned the gold standard.
Fundamentally, what is weird here is the elephant in the room this discussion ignores. Interest rates are implausibly low. The negative TIPS yield is just the natural result of reasonable inflation expectations, algebra, and Treasury yields that just a few years ago would have been assumed to be a typo. If you are looking for crazy, look there.