Inverted yield curve is a hot topic of conversation among the investment community of late. The yield curve is a measure of slope between different maturities on the Treasury curve. The slope is typically positive as shorter maturity bonds have lower yields and longer maturity bonds have higher yields with an expectation that inflation will rise in the future. On the other hand, an inverted yield curve is often a harbinger of slowing growth or recession in U.S., and it happens when long-term yields fall below the short-term yields. This generally occurs when the market is cautious about the future growth and inflation, while the Fed continues to hike the short end of the curve via the Federal Fund rate. The disagreement between the market and the Fed on growth and inflation then may result in an inverted yield curve.
Was inverted yield curve a strong predictor of U.S. recession in the past?
Duration, depth and breadth of the inversion are all important factors in assessing the weight of an indicator. A partially inverted yield curve or a one-day inversion of a narrow band by no means inevitably means an ensuing recession. For example, at the end of 1994 and the beginning of 1998, an inversion only happened for one day or even only intraday and it reversed to a normal spread almost instantaneously. Besides, the curve inversion in the 2H of 1998 could be seen as a false positive since no recession began until nearly 3 years after the inversion started and over 2 years after it ended. In fact, the 2 to 10 year spread only stayed inverted 3 months and was back to positive in Sep 1998 despite the 2 to 5 year spread kept inverted until the end of the year. Nevertheless, for most of the times in history, the front-end (e.g. 2 to 5 year spread) inversion usually was a prologue to a full spread inversion of the yield curve as typically the 2 to 10 years spread inverted sometime later and then followed by the 10 to 30 year spread. When an entire curve dropped into inversion, a recession always followed within 12 months to 24 months.
Figure 1. Recessions always followed with 12 months to 24 months when full inversion occur
How’s going this time?
The front-end of yield curve has recently inverted since 5 Dec, with 2 to 5 year and 3 to 5 year spread went negative of 1.2bp and 1.4bp respectively, while the 2 to 10 year spread also narrowed to 12bps from over 50bps in the beginning of this year. The trend looks ugly and a potential risk of U.S. recession in 2020 is on the radar. We are also cautious on US growth over the next 2 years, but we would like to highlight that the yield curve of today was affected by factors that were not seen in the past:
i) The global quantitative easing after the 2008 financial crisis made Fed to purchase significant volumes of Treasuries at longer maturity dates, meaning there are non-fundamental forces at work;
ii) The 2 and 3-year gross Treasury issuance has increased by nearly 30% YTD as well as the long-end 30-year Treasury of 20% YTD. In contrast, the 5, 7 and 10 year Treasury issuance was up by merely 8% only. As a result, a bond short squeeze may have been triggered by a rush of traders to rebalance their excessive short portfolio, making the spread narrows faster than actually the fundamental forces suggested.
Figure 2. U.S. Gross Treasury Issuance increased nearly 30% for short maturities.
Source: U.S. Treasury
Overall, we believe it is still too early to conclude a U.S. recession will come in 2020 as the Nominal Treasury yield curve is only inverted in the front-end in the moment, not to mentioned that the real (inflation-adjusted) yield curve on the front end is still positive. More importantly, given with the increasing market concern, there is still opportunity for the Fed to react to the indicator and thereby change the future. On the other hand, if the entire yield curve is really inverted this time, the coming recession could be much severe given that the current interest rate is much lower than our previous recession, leaving little room for the Fed to act.