Morgan Stanley Goes Rogue, Is Bullish On Bonds Because Frankly, Consensus Is ‘Usually Wrong’

As you know, the ongoing debate about where 10Y yields are heading and about what the bond selloff presages for equities is, well, it’s ongoing.

And when I say “ongoing” I mean it’s devolved into a veritable obsession. Over the weekend, Goldman “stress tested” the economy for the impact of a rate shock that would theoretically drive 10Y yields to 4.5% by the end of the year. That note came a week (give or take) after they upped their year-end forecast for 10Y yields to 3.25%. Other banks have followed suit.

Part of Goldman’s stress test involved projecting a 20-25% decline in U.S. equities, a precipitous dive that would feed through to the real economy by way of tighter financial conditions.

One thing to note about this whole debate is that last year, consensus was also overwhelmingly bearish USTs and we all know how that turned out. So it’s at least worth considering whether everyone might be wrong.

When you think about that possibility, it’s tempting to say something along the lines of this: “well, if you look at supply forecasts tied to financing Trump’s fiscal stimulus and tax cuts, and if you look at the possibility of foreign demand waning just as the Fed lets the balance sheet rundown and if you consider Fed hikes and inflation picking up and a rebuilding of the term premium and … etc. etc.”

While that’s all fine and good, remember that there were plenty of seemingly good reasons to be bearish on bonds headed into 2017 too. Maybe the list wasn’t as long and maybe the reasons weren’t as compelling, but there was a rationale.

Anyway, Morgan Stanley is out with a highly amusing contrarian take which is definitely worth highlighting in light of recent events and the veritable deluge of bearish bond calls.

“History has shown that consensus estimates for Treasury yields are usually wrong [and] everyone understands that accurate point forecasts rarely occur,” the bank’s Matthew Hornbach writes, before noting that “either the consensus is wrong in terms of direction or, when it has the direction correct, the consensus is wrong in terms of timing.”

He then moves on to detail what we noted above – namely that last year was a straight up disaster in terms of everyone getting it wrong on yields:

Last year, the consensus was wrong on direction. 10y yields started 2017 at 2.45% and consensus saw yields ending that year at 2.70%. Instead, 10y yields finished at 2.40%.

In 2018, folks have so far been right on the direction, but have been so far off on the timing that one wonders if anyone actually has a clue. Here’s Hornbach again:

This year, so far, the consensus is looking right on direction, but way off on timing. In response, forecasters have started moving higher their year-end targets. Since the year began, the Bloomberg consensus for 10y yields has increased by 15bp to 3.03%.

We’d be remiss not to note that this is the exact same dynamic we’ve recently pointed out with Tom Lee’s Bitcoin forecasts. That is, if you get the direction right but you are wildly off on the trajectory, then were you really “right”? Recall this excerpt from a recent piece we did on that and I think you’ll see the parallel:

If you predicted $25,000 from $4,000 over a five-year period and two months later prices had gone to say, $5,500, well then maybe you’re onto something. But there’s a sense in which you look just as clueless if you call for an asset trading at $4,000 to hit $25,000 in five years only to see it go to $20,000 in less than two months, as you would if you called for it to hit $25,000 in five years and it subsequently went to zero within 60 days.

Sure, you got the direction right, but that move from $4,000 to $20,000 in two months certainly seems to suggest that you were so wildly wrong when it comes to the trajectory that as far as whether you know what you’re talking about is concerned, it might as well have just gone to zero.

Anyway, getting back to yields, Hornbach isn’t surprised that everyone is suddenly adjusting their targets. To wit:

Every time the bond market moves dramatically and unexpectedly higher in yield, the consensus forecast plays catch-up. Exhibit 3 shows the 3-quarter ahead consensus forecast for 10y Treasury yields over the past decade alongside 10y yields themselves. The exhibit also shows that,even after yields stabilize, the consensus forecast continues to increase – likely reflecting bearish sentiment and recency bias in the wake of a large sell-off. We don’t feel compelled to join that crowd yet.

There you go. Morgan isn’t quite ready to jump on board. Rather, here is their take:

We Like The Long End.10y Treasuries are flirting with levels last seen in early 2014 while 10y yields elsewhere are still significantly lower. Global yields are beginning to roll over, led by JGBs threatening local lows. We think the bell has tolled for the best of the bear market in longer-duration bonds.

They’ve got a lengthy explanation for who’s going to absorb all of the new supply from Treasury, but that’s the bottom line from their perspective.

We’ll leave you with what is – by far – the best line from the note:

The wisdom of crowds has yet to grace itself on bond yield forecasters