We continue to believe curves will be flatter by year-end, which suggests that Treasury yields haven't yet peaked. Among the reasons are

  • An under-pricing of 2019 Fed interest-rate hikes,
  • Uncertainty about the effectiveness of fiscal stimulus, and
  • The impact of additional supply being auctioned.

Should curves continue to flatten, 10-year Treasury yields could breach 3% by year-end. The beta between 10-year yields and the two-year/10-year curve is about 0.50. If our forecast for another 35 basis points of flattening is realized, 10-year yields should climb about 18 basis points, which would mean about 3.1%, which is quite close to some technical resistance levels.

For this analysis we didn't include the period of the late 1990s "productivity miracle" and most of the quantitative easing (QE) period earlier this decade. The late 1990s saw yields rise above trend, while curves didn't significantly steepen as the Fed was on hold but growth increased. During the QE period, yields were depressed thanks to the Fed's large-scale asset purchases and the European sovereign debt crisis.

The long bull market in Treasury yields may be over.

For the past 30 years, cyclical selloffs in 10-year yields tended to peak two standard deviations above the long-term trend. The two-standard-deviation level currently stands at 2.81%. This is the first time since at least 1986 that yields have been above this two-standard-deviation mark.

We still question those who believe the break in this trend foreshadows a turn toward significantly higher yields, but it does suggest the old trend may be coming to an end. We think a more likely scenario is a multiyear, if not multidecade, broad range for Treasury yields, perhaps centered close to current levels.

  • Reasonable growth,
  • Low unemployment,
  • Inflation creeping higher,
  • Fiscal stimulus (albeit modest) and
  • Growing Treasury supply

.... combine to create a compelling case for higher U.S. rates. We remains skeptical yields will trend higher, but part of this may be our own recency bias. Over the past decade, each time it seemed yields might break out higher, weakening economic growth or other exogenous shocks brought a flight-to-quality bid into Treasuries.

Let's assume we're wrong and yields are biased to increase over coming quarters — we would still expect curves to be flatter than they are today. Secondary effects could also cause angst for fixed-income investors, as well as corporations and investors in other risk assets.

In the minutes from its Jan. 31 meeting, the Fed cited "substantial underlying economic momentum" and that inflation continued toward its target. Given that such trends are well underway with little obvious impediment, it appears the market may be underestimating the probability of rate increases beyond 2018.

Three hikes are basically priced in for calendar 2018. The FOMC has scope to upgrade its assessment of the economy and may reiterate that "further" hikes are coming. This should continue to affect expectations for 2019 hikes. At the moment, fed funds futures are pricing in less than two increases next year — a signal for three should mean a noticeable selloff in the front end of the Treasury market.

Expected Fed activity tends to have an outsized impact on the shape of yield curves. If the market were to price in additional hikes for 2019 given improving economic prospects, yield curves should flatten further. Historically, as curves pancake, yields tend to peak.

The bear flattening that accompanies removal of monetary accommodation is a known phenomenon. With the two-year/10-year Treasury curve still having more than 60 basis points to flatten, there may be support for the view that there's more room for the 10-year to sell off in the intermediate term.

Ira Jersey is BI Chief U.S. Rates Strategist; Aleksandr Nozhnitskiy is a BI Credit Associate Analyst.