Welcome back. When I took this job, my friend Jonathan Guthrie said that the professional risk was that I would just end up writing, “boy, interest rates sure are low!”, in slightly different ways, day after day. He has not turned out to be entirely wrong. On the other hand, there do seem to be a lot of ways to think about low rates, and which way you think about them does seem to matter to how you see what will happen next.
That said, here we go again.
It’s a mistake!
On Wednesday, I shamelessly anthropomorphised the bond market, trying to sort out what it was “telling us” about the outlooks for monetary policy and the economy, in light of the fact that 10-year yields didn’t seem bothered by a hot inflation report, but 2-year yields snapped to attention. Here is what I came up with:
Inflation is looking mostly transitory, but there is enough of it, and short-term growth looks good enough that the Fed will have to taper bond purchases in a little while, and raise short rates after that, so short yields need to nudge up. BUT these Fed moves will make it all but certain that inflation doesn’t get out of control, and indeed, growth isn’t going to be that great from here on out, so no need to bid up long bonds on inflation risk.
Not everyone agrees. I was particularly struck by this alternative view from Thomas Tzitzouris of Strategas:
Like a surgeon that always wants to cut, the Fed, over the last 40+ years, has consistently chosen to tighten more than the bond market said was warranted. So betting odds would suggest that once inflation enters an uncomfortable zone, the Fed is going to respond by raising short rates more than they should, and thus forcing long rates lower, all the while scratching their head and uttering gibberish about why the move in long rates doesn’t make sense.
The Fed will kill the economy by tightening, despite supine long yields telling them, loud and clear, that they don’t have to. Wednesday’s move up in the 2 and the flat 10 was “the bond market’s way of saying ‘policy mistake is coming’”.
I wrote to Tzitzouris to ask why we had to anticipate a mistake. Can’t it just be that there is enough inflation for a little short-end tightening, confidence that the Fed will ultimately control inflation, and a modest, “new normal” level of long-term growth — all justifying a flattish curve? He replied:
Historically, the Fed has regularly ignored the fact that they’ve proven impatient and artificially moved rates well above healthy levels on the front end and forced the back end lower, and then continued to raise rates into inversion [of the yield curve] . . . [Ben] Bernanke began to accept that the market had a better grasp on where neutral [interest rates were], and [Jay] Powell appears to fully embrace the notion that no amount of PhDs can match the wisdom of the cowboys in the trading pits (or over the phones in the case of bonds). Still, the market thinks the Powell Fed is no different from prior Feds, and the market is saying Powell will eventually tighten until something breaks.
I’m not sure what I think about this. But what is very striking is the idea that the market thinks the Powell Fed may overdo it, despite Powell trying as hard as he can to tell us he definitely won’t do that (as he did again on Wednesday). He’s a prisoner of history.
No, it’s about liquidity
According to Michael Howell of CrossBorder Capital, the yield curve isn’t telling us anything about what central banks might do. It’s telling us about something they are doing already: pulling liquidity out of the system. Here is a chart from CrossBorder that shows how the pace of liquidity injections in Europe, China and the US has fallen:
This is a chart of indices, and takes a little explaining. A level of 50 represents the five-year rolling average liquidity growth in each jurisdiction, measured by the changing size of central bank balance sheets along with other factors that release money into/absorb money from the economy. The index shows deviation from that mean, with 20 units representing about a standard deviation move. So with the Covid response, the pace of liquidity introductions hit highs nearly two standard deviations from the mean, but now the pace is dropping back.
In Europe and China, this deceleration comes down to a straightforward slowdown in central bank asset purchases. In the US, asset purchases continue at $120bn a month, but two other factors play in.
For one, the Treasury’s bank account at the Fed (the “General Account”) is getting smaller. Basically, the Treasury raised a lot of money and put it aside. Now it’s spending that money, putting it into the economy. On the other hand the Fed, worried about money-market interest rates going negative in a market awash with cash, is absorbing liquidity with reverse-repurchase operations, swapping its assets for cash overnight. The Fed says this is a purely technical operation. Liquidity watchers think it is reverse-QE at the short end of the curve (“tapering in disguise” as Eric Barthalon of Allianz described it to me).
Here’s how those two forces compare (Fed data, my chart; who else would make such a crappy chart?):
Since early April, reverse repo balances have risen by almost $800bn. The General Account has fallen by just $250bn or so. US liquidity growth is slowing.
What does all this have to do with the yield curve? The view of Howell at CrossBorder is that the financial system today is not so much for financing — funding new companies and projects — as for refinancing, given the huge amounts of corporate and sovereign debt in circulation and the perpetual need to roll it over. When liquidity starts to recede, refinancing debts looks harder, and leveraged balance sheets start to constrain activity, which “increases uncertainty, limits economic growth and potentially causes volatility to spike higher”. In such an environment, Treasury bonds look pretty good.
“If you flood the system with liquidity, the demand for safe assets goes down, because you want risky assets. When liquidity falls, everyone wants the safe asset — which explains very well why the yield curve moves pro-cyclically with the liquidity,” he says.
When liquidity falls, people want Treasuries, which pushes long yields down and flattens the curve. Howell illustrates the point with this chart of the 10-year-5-year yield curve plotted against his US liquidity index (with liquidity moved forward by nine months):
The higher demand for safe assets is not meeting with greater supply. Net of Fed purchases, the Treasury is not issuing tons of long bonds right now. Here, from the Dallas Fed, is the amount of Treasuries outstanding in the private sector:
This is all a little involved. Andrew Smithers, however, wrote in response to Wednesday’s letter, making what I think is broadly the same point: one has to think about rates in terms of supply and demand, not what bonds are “telling us” about the economy.
“Prices are determined at the margin,” he wrote. “The yield [curve] is not ‘the bond market’s view’, it is the view of bondholders at the margin.”
One good read
I am not a religious man, but I am interested in the ancient world, which led me to Sarah Ruden’s new translation of the four gospels. Rather than modernising the text, she tried to preserve the feel of the original Greek text, with its Aramaic and Hebrew interpolations. The result is a weird, messy, deeply foreign but moving book.