For most of 2017, interest rates followed patterns that have been familiar over the past couple of decades. Estimates of term premia remained low, or negative. Inflation compensation from TIPS was very low. Even as the FOMC tightened the federal funds rate 75 basis points, long-term yields actually dropped. This is reminiscent of former Chairman Greenspan’s conundrum and has been a repeated pattern over the 21st century so far.2
Things have turned around sharply in the last two months, and ten-year yields are up to 2-3/4 percent. Some part of this is an upward revision to policy expectations. The large fiscal stimulus should boost neutral real rates by at least 25 basis points. Financial conditions remain accommodative, notwithstanding the recent slide in stock prices, and the dollar has depreciated, both further raising short-run neutral rates.3In addition, some signs that inflation is moving up make it more likely that the Fed will continue tightening policy than appeared to be the case last year. But these factors affect the near-term outlook for policy, not long-run r
star and pi-star. So it doesn’t seem reasonable to account for the move-in rates with policy expectations alone. Instead, I would argue that term premia have moved back up, faster than had been expected.
The rise in term premia is more about “animal spirits” in the bond market than any fundamental factors, and the talk of a regime shift in fixed income markets is premature. The unwinding of central bank purchases should boost term premia, but this is not a new factor. There has been some discussion of the effects of Treasury issuance on term premia, as distinct from its effect on short rates, and I see this as a very small factor, that is a matter of a few basis points.4 Meanwhile, the forces that have lowered term premia are still there: fear of deflation, financial instability, and the zero lower bound are much greater risks than a repeat of the Great Inflation of the 1970s. I think that a plausible steady-state term premium is close to zero but positive⸺perhaps 20 basis points or so for the ten-year term premium⸺and we have moved towards that quickly over the last few weeks.
Unlike in previous years, survey expectations of monetary policy and the SEP are broadly in agreement. Raw money market futures quotes imply about 65 basis points of federal funds tightening in 2018, and just 25 basis points in 2019⸺quite a bit less than what appears to be the consensus. Part of this is because surveys ask for the modal forecast whereas futures markets, under risk neutrality, deliver the mean forecast.5It seems
1 Jonathan Wright is a Professor of Economics at Johns Hopkins University and a Senior Adviser to Monetary Policy Analytics. 2 For a more complete discussion, see Hanson, S.G, Lucca, D. and Jonathan H. Wright (2017): Interest Rate Conundrums in the Twenty-First Century, New York Fed Staff Report 810, available on the web at https://www.newyorkfed.org/research/staff_reports/sr810.html 3 The dollar depreciation is mysterious given that US interest rates have climbed more than abroad, but interest differentials have a poor long-term track record in explaining exchange rates.
4 Treasury has stated an intention to keep weighted average maturity around its current level, ending the trend to a longer maturity of debt stock.
5 The modal outcome is the most likely path for policy; the mean outcome weights different possible paths by their respective probabilities.
reasonable to suppose that the mean forecast for interest rates is below the modal outcome at present.6 Even allowing for this, the futures quotes continue to suggest a slightly negative term premium at the front of the curve. The term structure model maintained by the New York Fed puts term premia a bit below zero across the curve. A simple comparison of ten-year forward rates to consensus views of long-run r-star and pi-star implies that ten-year forward term premia are very close to zero.
Going forward, if the Fed continues tightening at a pace of about 25 basis points per quarter and term premia remain around zero, the nominal Treasury yield curve will be almost flat in the middle of next year. Near inversion of the yield curve will get a lot of attention and some policymakers might view it as a reason to hold off on rate hikes. It’s not an argument to be lightly dismissed, because all recessions over the last 50 years have been preceded by a very flat or inverted yield curve, although with somewhat variable lags. Nonetheless, an inverted yield curve in a world where the ten-year term premium appears to be around zero has to have a very different meaning from an inverted yield curve with (say) a 4 percent term premium. In an environment with term premia hovering around zero, any time that the fund’s rate is above its long-run expectation, the yield curve should invert, and that should, of course, happen about half the time! In contrast, in the past when term premia were big and positive, an inverted yield curve signaled policy that was very restrictive relative to long-run expectations⸺either a deliberate disinflation, or a policy mistake. All in all, if a flat yield curve in a context of near-zero term premia in 2019 comes to pass, I don’t think that it should in itself be interpreted as a strong signal of an imminent downturn.