Term premium and the yield curve

There has been a notable flattening in the US Treasury yield curve since the last FOMC announcement. 2 & 5 year Treasuries have sold off more aggressively, with yields around recent highs.  10 year bonds have moved less, and at ~2.80% are still well below the recent 3% high seen at the end of last year.

Assuming the economy continues to track as the FOMC expects and we get closer to forward guidance becoming a reality, a key question will be whether the yield curve will continue to flatten.

This certainly was the experience at the beginning of the last tightening cycle from the FOMC, that began in 2004. At the beginning of this cycle, the yield curve from 2 - 10 years was about as steep as it is now (although the Fed Funds rate was at 1% rather than the zero lower bound).

There are a wide range of differences between this time around and 2004, with the unwinding of QE, depth of recession and trajectory for growth and inflation particularly important.

One area I want to focus on is the term premium in 10 year treasuries. This is typically thought of as the premium investors are paid for holding bonds of longer maturities above what would be earned from rolling over bonds of short maturities.

At present, the term premium is still very low and has only just recently turned positive (according to BIS calculations). Indeed, for much of 2012 and 2013, rather than being compensated for interest rate risk, investors were paying a premium to be in longer maturities (vs rolling shorter maturities).

A central question as to whether the yield curve flattens forward guidance come closer to reality depends on whether the term premium remains very low or moves back towards recent history of 50-100bps.

One piece of fairly recent research from the FRBNY, which can be found here, suggests that history suggests the commencement of tightening doesn't push term premia significantly higher. That said, there is the caveat that current circumstances are quite different given the current size of the Federal Reserve's balance sheet.  To be sure, it was a shift on expectations on the size of the Fed's balance sheet that probably helped push the term premium from -100bps back towards zero.

So for now markets seem happy to let the non-zero parts of the yield curve flatten a little. But I think it is fairly limited as to how far this can go given the risks around the term premium.

While a flattening or steepening yield curve has lots of implications for growth and financial markets, in the commodities world, the clearest link is to gold, for which real interest rates are particularly important.

While rising interest rates is a negative for gold in the absence of a weaker dollar, a flattening yield curve would mean the outlook is perhaps less bearish than some are projecting.

Gold can have is ow momentum and has seen several paradigm changes in the way it moves against bonds and the USD in the last 30 years (see this post).  More recently, however, these periods where gold has diverged from bonds have tended to be short lived.