This commentary was written by Jonathan Wright, an LH Meyer senior adviser. He is currently Professor at Johns Hopkins University.
Since just before the election, ten-year Treasury yields have climbed 55 basis points and futures rates for the end of 2018 have gone up 70 basis points. The increase is most pronounced in forwarding rates about three years out. The rise in rates reflects stronger incoming data and an expectation for fiscal expansion. But part of it is also some normalization in term premia.
The size and type of fiscal stimulus that is coming remain quite unclear, but the most likely outcome is surely a material fiscal expansion. The latest New York Fed primary dealer survey, a natural benchmark for market expectations, predicts a fiscal stimulus of 0.3 percentage points of GDP in 2017 and 0.8 percentage points in 2018. A stimulus of that sort will push short-run r-star up and lead to a firmer path of policy. The impending stimulus complicates the FOMC’s economic projections and communication strategy, but the Committee will wait for Congress to enact a fiscal stimulus before actually altering policy. However, the rise in rates means that markets have pre-emptively responded to the prospect of fiscal expansion and already offset part of its effects on aggregate demand.
While the fiscal expansion could also boost long-run r-star, that effect is likely to be very small. The tax code is inefficient, and there are surely ways of reforming it that could boost productivity growth. But tax reform is politically hard. Scattershot reductions in tax rates are politically easier, and hence more likely.
According to the New York Fed primary dealer survey, the modal outlook before the November FOMC meeting was for five 25 basis point rate hikes in 2017 and 2018. Ahead of the December meeting, respondents added in one more 25 basis point increase. The primary dealers however saw this as just a change in the timing of rate normalization. Perhaps surprisingly, the average expected federal funds rate over the next 10 years rose only 2 basis points. That’s qualitatively consistent with the increase in forwarding rates being concentrated about 3 years out—not in distant-horizon forward rates. It is well known that survey expectations can be somewhat too inertial, but even allowing for this, the rise in yields has been larger than a reasonable assessment of the increase in monetary policy expectations. Laubach (2009)1 considered various methods for calibrating the effect on interest rates of a one percentage point of GDP increase in the budget deficit and concluded that an impact of roughly 25 basis points was a good benchmark calibration.
As the rise in Treasury yields and interest rate futures rates since November has been too big to be explained by a revision to monetary policy expectations alone, this leaves a rebound in the term premium as a substantial part of the explanation. For most of 2016, measures of term premia were exceptionally negative. I interpreted these negative term premia as reflecting bonds being a hedge against the risk of the economy becoming stuck in a deflationary spiral at the zero lower bound. That is still a risk, but it is not
clear that it is the predominant risk anymore. Even a modest increase in short-run r-star makes a new brush with the zero lower bound less likely. When term premia were at their most negative, it seemed reasonable to suppose that they would eventually normalize—that is indeed built into most term structure models. That normalization process seems to have begun quicker than one might have expected.
As of January 10, the term structure model maintained by the New York Fed puts the ten-year Treasury term premium very close to zero. Comparing survey expectations with futures/OIS rates suggests that shorter-maturity term premia remain perhaps slightly negative, but are very close to zero. This still leaves term premia very low by historical standards. I would expect term premia to rise a bit further in the future, perhaps helped by some eventual shrinking of the Fed balance sheet. But absent a very big change like the risks facing fixed-income investors, and especially the inflation outlook, I do not see term premia rising back to the levels that were normal in the 1980s and 1990s. Although a lot has happened in fixed income markets in the last couple of months, the talk of a regime shift is premature.
Finally, in 2016, there was some evidence that market participants’ density forecasts for interest rates were skewed in the direction of lower rates: Modal interest rate forecasts were above mean forecasts.2 That’s a separate point from term premia, but it is relevant for reconciling interest rate futures/OIS quotes with survey forecasts, which generally report the most likely outcome. It might seem intuitive that in the new monetary policy environment, the risks are more symmetric, encompassing meaningful odds of either a return to the zero lower bound or a more brisk pace of tightening. However, that does not show up in the December primary dealer survey, where the density forecast continues to be skewed in the direction of lower rates.