Over the weekend, we published an analysis from Citigroup looking at how long after the yield curve inverts do investors "have to worry." The results were interesting: as Citi wrote, while sometimes inversion provides a timely signal for the economic cycle a la 2000, "where Professor Curve predicted almost the ding-dong high in the SPX", other times, like the 2006 inversion, dished up 7 months of pain for equity bears, with 18% further upside for the SPX. The same occurred for the 1989 episode where equities continued to rally 22% into the 1990 recession.
Whatever the timeframe between inversion and subsequent events, however, the curve first has to invert. So when will that happen. One bank provided a surprising answer earlier this week, when Morgan Stanley forecast a "completely flat" yield curve around the time of the FOMC's September meeting.
Now, in its 2018 rates outlook, BMO's Ian Lyngen and Aaron Kohli have unveiled a far more aggressive forecast, warning that there is a "risk of an inverted 2s/10s curve as early as the March meeting – if not, then by June."
Here are the details, as excerpted from the report:
As we contemplate our Treasury market call for 2018, 10-year yields are at 2.35% -- not dissimilar to the end of 2016 (2.44%), 2015 (2.27%), and 2014 (2.17%). While the yearly figures suggest that there has been a gentle upward grind in yields, in fact the path has been more volatile and along the way 10s managed to reach 2.63% during 2017 on the back of Trump-o-nomics euphoria, only to retrace to as low as 2.016% as geopolitical risks combined with the realities of lowflation. Our primary focus for 2018 will be on the shape of the yield curve and we expect that as the Fed’s hiking campaign soldiers on, the most material risk is that monetary policy inverts the 2s/10s curve more quickly than anticipated. We have 10-year yields penciled in as closing 2018 at 2.40%, but not before a test of the bottom of the recent range (2.00%) and the risk of an inverted 2s/10s curve as early as the March meeting – if not, then by June.
While we’ve been focused on this risk since September ‘17 when the Fed signaled its intension to push forward despite realized inflation, it has become increasingly consensus – a dynamic that always gets us nervous, but we’re nonetheless convinced of the fundamentals behind the call. If anything, the newfound popularity of the forecast leaves us open to an inversion earlier in 2018, which will subsequently be followed by a resteepening.
* * *
We are certainly cognizant that forecasting 10-year yields will end next year at 2.40% after spiking in the first quarter only to rally throughout the summer months is consistent with the seasonal patterns and doesn’t represent an exciting break from the norm. As we watch the market trade a version of effectively the same tax reform deal for the fourth time and 10-year yields unable to even touch the 2.64% peak achieved immediately in the wake of Trump’s election, we find ourselves content to continue playing the range for the time being.
As BMO adds, "the inflection point will be a result of either the real economy falling far short of the Committee’s expectations or a moment when the Fed effectively blinks via a pause in the tightening campaign. The outcome of either event will be a flattening out of the OIS curve and the resteepening of 5s/30s. If 5s/30s is the coalmine canary of monetary policy expectations in the current environment, then the bullish steepening that followed a dovish read on the November FOMC meeting Minutes is a useful roadmap for the return of term premium."
BMO sees one case in which the curve steepens:
the curve will steepen out and term-premium will return if the Powell-Fed sees unexpectedly high inflation and doesn’t step-up the hawkish rhetoric – implicitly or explicitly signaling a willingness to let the economy ‘run hot’ for a time. An alternative method for getting term-premium back would be to alter the Fed’s communication strategy so drastically as to leave the market uncertain as to how the Fed will address the developments in the real economy – said more directly: dropping the dot-plot and forward rates guidance.
However, as the authors concede, "we seriously doubt this is in the cards and only offer it as an example of the degree of change in the Fed’s behavior that it would take to reintroduce volatility in the longer end of the Treasury curve (10s and 30s)."
Ultimately, the conclusion is simple: "whenever the Fed is actively engaged in raising rates the curve simply continues to flatten even with an acceleration of inflation." And if BMO is right, the Fed's rate hike cycle has roughly 2 more quarters to go.