While most financial professionals would enjoy nothing more than to trade bitcoin all day long - after all it only goes up in the long run (for now), while bringing insane volatility with it - the reality for most is that they are confined to such boring, established legacy instruments as Treasurys, which rarely have intraday fireworks similar to those in bitcoin. There are rare exceptions of course, and as we showed earlier today, someone made roughly $60 million betting on the sharp yield curve steepening we have observed in the past 2 days following a record pile up of long net specs in the 30Y future.
There is an additional benefit of trading boring , old TSYs: one gets to indirectly at least, determine the time until the next recession, because the flatter the yield curve, the closer the recession, no matter the mitigating bullshit so-called experts will spout to keep confidence high.
So where do we stand right now? According to a recent analysis by Goldman, which expects no less than 8 rates hikes amounting to 200bps in Fed Funds futures by the end of 2019, the bank expects "rising US policy rates and term premium to gradually drive 10y rates higher, with the long end also rising more than market pricing. We also expect further US term structure flattening." Just not too much; in fact in keeping with the whole Goldilocks theme, "just the right amount of flattening", to wit:
For much of 2017, the term structure flattened because long-end rates fell, not because short-end rates rose. Indeed, by early September, US 10y rates had fallen 36bps, while US 2y rates had risen only 9bps on the year. But since then, US 2y rates have risen more than 55bps, while US 10y rates have increased only 29bps. In 2018, we expect 2y rates to be the driving force in flattening, as was the case in previous late cycles; CFTC positioning also shows investors are more bearish US 2y than 10y rates.
Ok fine, but with the 2s10s flirting with 60bps even after today's violent selloff, there is only so much the Fed can hike rates - just over 2 times to be exact - before the curve inverts unless the long end move higher in a parallel or steepening shift. Goldman admits as much, noting that "we would expect a flatter yield curve to signal rising recession risk and a warning for risky assets, in line with history." But - to keep its clients happy, and buying all the accumulated inventory for sale by Goldman's prop desk, Goldman does not forecast a recession in 2018. Instead, "we are OW equity and UW bonds over both 3 and 12 months in our asset allocation." Finally, Goldman's rates strategists forecast the US 10y-2y slope at 35bps at 2018YE.
But doesn't that mean an imminent recession?
Here readers can decide on their own: below is the Goldman chart showing the 20/80 percentile corridore of all previous recessions, and where the 2s10s traded in that time, as well as the average as the US headed into a recession.What it shows is that contrary to what optimists may claim, the curve is flat enough to be caught by the upper end of the confidence interval, and that should it continue flattening at the current pace, a recession is indeed likely in roughly 12 months, all else equal.
Finally going back to an analysis we showed three weeks ago, Citi calculated that assuming a runrate of 40bps of flattening per year, then we could see a flat curve within 18 months (Figure 3, left). So when Citi asks "should we be worried?" it answers that, rightly or wrongly, market participants with grey hairs would preach that all is well until the curve begins to invert. Which, however, is not the full story as Citi explains:
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. However the 2006 episode of inversion dished up 7 months of pain for equity bears, with 18% further upside for the SPX. Ditto for the 1989 episode where equities continued to rally 22% into the 1990 recession (Figure 3, RHS). For now, we’re comfortable with the flattening dynamic with regards to other markets but would become increasingly cautious as the curve approaches zero.
In other words, even when the curve fully flattens and ultimately inverts, it is unclear how much longer risk assets have before acknowledging that a recession has arrived: in the Y2K case, inversion marked the top; on other occasions the stock market may rally for many painful months, rising as much as 22%, before admitting defeat.
And finally, one should also remember that the current cycle is unlike anything ever observed before thanks to some $15 trillion in excess liquidity, which makes all such historical comparisons moot and generally a waste of time.