If you're one of the folks who have tried to take a contrarian approach and lean against inexorable curve flattening as opposed to leaning into it harder than a Texas rapper on purple Sprite, well then you've probably had a rough go of it.
Take Monday for instance, which finds the 5s30s eying 70bps:
This has become a veritable obsession for markets as everyone scrutinizes the tea leaves on a near-hourly basis in the hope of catching a whiff of a looming recession. Much more on why long rates refuse to move here.
As ever, the trend is your friend, something CTAs know all too well. According to Deutsche Bank's one-month rolling betas, CTAs are still in bear flatteners – they've been short since late last month and in flatteners since midway through August.
Meanwhile, in what has to be one of the more amusing similes ever employed by the sellside, Morgan Stanley is out saying that “The flattener is like Bitcoin.”
How's that, you ask? Well apparently because you should just not question this trade but rather hold your nose and ride it some more. Here's MS:
With the consensus CPI report in hand, we scale back up our UST 2s30s flattener. We've made valuation arguments over recent weeks for the flattener:negative term premium in the front end and positive term premium in the long end of the yield curve. With the Fed hiking rates amidst below-target inflation,now is as good a time as any to harvest those term premiums. Trading the flattener should feel very much like trading Bitcoin: you're meant to buy every dip, if you see the value proposition, even though you've already missed what seems like a big move.
And as if that wasn't enough, they're even throwing in some Fibonacci projections for dat ass:
How far can the flattening go? We certainly believe the curve will eventually become completely flat if the Fed continues to hike rates.From a tactical standpoint, we think the 2s30s curve could flatten from its current level in constant maturity terms (1.09%) into the 0.80-1.03% zone as defined by Fibonacci projections (see Exhibit 3) into yearend. Those projections come from multiplying the distance the curve steepened in 4Q16 by Fibonacci percentages and subtracting the product from the point when the steepening began.
We can also look at Fibonacci retracement levels from the curve steepening that occurred during the financial crisis (see Exhibit 4). The curve flattening that began in early 2011 encountered resistance around the 38.2% retracement level,and then at the 50% retracement level,and then at the 61.8% retracement level. Most practitioners ignore the 76.4% retracement level, thinking that a break of 61.8% means the market will retrace the entire move. But we can still use it as a reference point. The 76.4% retracement at 0.86% on the constant maturity 2s30s curve fits with the idea that the curve will find resistance to further flattening in the 0.80-1.03% zone.
