I read an article recently that noted the 65-75bp rise in Treasury yields over the last year or so, and sought to explain, through a labyrinthine line of reasoning/model, that most of the rise was due to the “reflationary” trade, with the Fed hopelessly behind the curve. The model the author used depicted inflation expectations as being fairly directly tied to the rise in inflation outcomes that we’ve seen: headline inflation has risen from below 1% at the middle of last year to 2.4% year-over-year ended last month.
This approach was, at one time, fairly standard. Since there was no way to directly observe inflation expectations, people measured real rates by taking current interest rates and subtracting trailing 1-year inflation, reasoning that recent inflation is a good proxy for expectations.
Indeed, you will still see some economists and bloggers referring to the “recent decline in real rates” that has happened since headline inflation has risen about 250bps since mid-2015 (see chart, source Bloomberg) while 10-year rates are approximately unchanged over the same time horizon.
With this framework, economists would say that real interest rates have fallen precipitously and are now roughly zero, whereas two years ago they were over 2%.
Of course, that old way of doing things is nonsense today. Because past inflation is highly influenced by changes in energy prices (oil prices bottomed in early 2016), trailing inflation is in fact a pretty poor measure of longer-term inflation expectations, and we no longer need to rely on this method because we can directly observe real interest rates, and to some extent market measures of inflation expectations.[1]
Here are the current levels, along with 1-year and 2-year changes, in real rates and inflation over the last one and two years (source: Bloomberg; Enduring Investments calculations):
So what has really happened to longer-term real rates and inflation expectations? Over the last two years, 10-year real yields have risen about 27bps, with roughly unchanged 10-year inflation expectations, producing a 25bp rise in nominal interest rates. Over the last year, those numbers are +38bps and +27bps, leading to a 65bp rise in 10-year nominal yields.
Those figures give the central bank tremendous credit for not being behind the curve. Over the last two years, core and median inflation has risen 0.3% while 10-year expectations have been stable. Over the last year, core inflation has fallen a bit (though that has a lot to do with the quirky plunge in telecom prices last month, which should be reversed this month) while median has risen about 10bps.
Still, there’s no panic at all in inflation markets. Real yields have risen only 16-65bps over the last two years, despite 75bps of rate hikes.
The Fed very probably is well behind the curve, but the market doesn’t think so. The bond vigilantes haven’t even begun to light their torches yet.
[1] Since market nominal interest rates are lower than they would be if the Fed had not bought a few trillion in securities, breakevens and inflation swaps are probably lower than true inflation expectations would be if the market was freely trading, but since at some point market rates will begin to anticipate the unwind of the Fed’s balance sheet we can’t really say for sure.