Welcome back. The week is about over, and I promise next week we won’t talk about inflation at all. I really mean it this time. But today we will.

Email me your thoughts:

The consumer price index came in a bit hotter than expected on Thursday, with core inflation at 3.8 per cent, but government bond yields didn’t move. In principle this is a little weird. Inflation is bad for bond prices, so it should drive yields up. But bonds’ indifference is not unexpected. Bond yields peaked back in March. Since then the story has been “look at the 10-year note, it says inflation will be transitory, everything is cool” or alternatively “look at the 10-year Treasury, it says that investors think inflation will be transitory, but boy are they in for a surprise, buy canned food and guns”.

I am closer to the former camp than the latter. Once again, most of the stuff that drove the index up in May was stuff that the pandemic crushed, including hotel rooms, or created bottlenecks in, such as cars. Capital Economics had a tidy chart of the hot categories:

All of that should be transitory. But we cannot quite relax. Three comments on why not.

First, not absolutely every category that is heating up can be dismissed as a natural consequence of the reopening. Housing costs (“owner-equivalent rent”) which rose at an annualised rate of more than 4 per cent was one example. It’s not a crazy number (“so far there is a normalisation, not a surge”, Strategas wrote in a reassuring note). But it makes me want to see what next month’s number looks like.

Second, there is a lot of price-insensitive demand for US sovereign bonds, which may prevent yields from responding to inflation fears. They are the most liquid of assets, used for all sorts of purposes other than maximising returns. They are a safe cash alternative and a form of collateral for just about everyone, everywhere.

An example. My former colleague Tracy Alloway, now at Bloomberg, had a nice article this week about increasing demand for US Treasuries at banks, which are required to hold a bunch of very liquid, safe paper. Collapsing yields on the other options have shifted bank demand toward Treasuries, and banks have bought hundreds of billions of them in the past year or so.

On top of that, remember that US Treasuries still yield a lot more than other sovereign bonds. Japanese bonds yield basically nothing. German ones have a negative yield. So if you have a safe sovereign bond allocation that needs filling, what are you going to fill it with?

And, oh yeah, on top of that, the Federal Reserve is buying $80bn of Treasuries a month. That is almost half of the net issuance over the past 12 months, on data from the Securities Industry and Financial Markets Association, or about 4 per cent of the outstanding stock of Treasuries. As the trader who tweets as Five Minute Macro summed it up:

I mean, if the 10-year Treasury yield doesn’t tell us much, what the hell am I doing for a living? But, anyway.

Lastly, we have a perfectly simple explanation of what is going on, which is that inflation fears’ effect on yields are being masked by falling real rates. Here are market-derived inflation expectations for the five years starting five years from now, plotted against the 10-year yield (data from the Fed):

To simplify, inflation expectations can go up while yields stay flat because the inflation-adjusted return that investors demand on the money, the real interest rate, is falling.

I wrote about real rates yesterday. Thinking about it since then, it occurs to me that a very low and falling real interest rate is hard to tell from investor nihilism (“the return on everything stinks, I’ll settle for anything with a tiny bit of yield, going to cash and waiting for something to happen will get me fired, is it drink time?”) but that is an issue for another hour.

The Basel Committee on Banking Regulation thinks that banks that hold cryptocurrencies should keep capital equal to the full value of those digital assets. In Basel-speak: “Capital [should be] sufficient to absorb a full write-off of the cryptoasset exposures without exposing depositors and other senior creditors of the banks to a loss.”

This makes perfect sense, and makes crypto a terrible business for banks.

It makes sense because cryptocurrencies (except for those tethered permanently to more stable assets, which Basel has excluded from the heavy capital demands) are wildly volatile. Bitcoin lost almost half its value in a few weeks in May for no apparent reason. A bank can’t go around putting leverage on something that behaves that way.

This is obvious, and is in addition to technological or criminal risks associated with crypto (“cryptographic key theft, compromise of login credentials and distributed denial-of-service attacks”).

But banks make basically all their money from leverage. Their return on assets is about 1-2 per cent, they lever that 10 times or so, arriving at a return on equity that is barely higher than their cost of capital. They look like they make a lot of money in the good times, but that’s an accounting illusion. Across the cycle it’s a pretty tough business. Assets that cannot be leveraged don’t fit the business plan, at least not at any scale that matters.

This is not a criticism of bitcoin or crypto assets generally. And it shouldn’t bother crypto-believers much. A key part of the pitch for crypto is that it will allow users to tell the government-controlled system of banking and money to buzz off. If that system wants crypto to buzz off too, well, everyone should be happy. Bitcoin didn’t move much on the news from the committee.

The Financial Times did, however, find a banker willing to say on background that the committee has it wrong:

This is a hilariously bad argument (“If you don’t let bankers smoke crack, who is going to smoke all the crack? The children!”). Yes, we do want crypto to be handled in some sort of self-contained system where, if it blows up, we don’t need to hold a mortgage-bond fire sale to fill the hole the explosion left behind. The interesting question is how that self-contained system should be regulated, taxed and so on.

The FT’s Martin Sandbu thinks there is no labour shortage, no widespread wage pressure and no sticky inflation. He backs his argument with plenty of data and sound logic. Read his column if you are feeling panicky.