Core inflation falls, a lot

Receive free US inflation updates

We’ll send you a myFT Daily Digest email rounding up the latest US inflation news every morning.

Good morning. Former retiree Bob Iger has been given a couple more years as “interim” CEO of Disney, and the ceiling on his bonus pay has been quintupled. This reminds us of an old Peter Arno cartoon depicting a bunch of ancient gentlemen seated around a long table, wearing wrinkled grins. The caption: “Then it’s moved and seconded that the compulsory retirement age be advanced to ninety-five!” Feel young at heart? Email us: robert.armstrong@ft.com and ethan.wu@ft.com.

That disinflation sensation

Although we passed the inflation inflection two months ago, the speed of inflation’s descent has remained in doubt. Yesterday’s CPI numbers offered reassurance on that front. Core inflation, which had seemed hot and sticky for six months running, rose less than an annualised 2 per cent in June. Goldman Sachs called it a “turning point”; for Standard Chartered, it was a “game changer”.

The details looked encouraging, too. Used car prices fell, finally relenting from a shortlived demand pop. Core goods prices declined slightly. Core services inflation rose a modest 0.25 per cent, in line with the pre-Covid average. The cooling in core services prices was helped by slower rent inflation. Newly-signed leases are trickling into the CPI data, which also includes old, peak-inflation leases. This transition takes time, so the economist Jason Furman uses private market rental data to simulate where core inflation will settle once CPI finishes catching up. On a six-month annualised basis, core CPI would be at 2.5 per cent if CPI shelter fully reflected private market rents.

A few notes of caution are needed. Next month’s report may not be so benign on the services side. Hotels and airfares, two volatile services categories, posted huge price drops, at 2 per cent and 8 per cent month on month, respectively. These will probably reverse. And other services categories such as vet visits, recreational services, car insurance and repairs are all still running hot (for a deeper discussion of car insurance inflation, listen to the Unhedged podcast).

But good news is good news. Most encouraging of all, inflation appears to be unsticking. The indices below measure, in different ways, how broadly entrenched inflation is. All three are pointing in the right direction, and one is even sporting a 2-handle:

The market was pleased. Falling yields carried stocks gently higher. Calmer inflation means rates don’t need to go as high, or for as long, as the market had expected. After the CPI report, the futures market lowered its probability estimate of a second additional rate increase this year (the one in July still looks all but guaranteed). The two-year yield fell 13 basis points, confirming the signal from the futures market.

But these were incremental moves, reflecting the incremental information contained in yesterday’s report. The data did not guarantee a soft landing, but suggested what the path may be. That is: decelerating job growth plus lower core inflation buys room for the Federal Reserve to go easier. The chance of that happening is on the rise, but things need to keep going right, so Mr Market is reserving judgment. As BNP Paribas analysts wrote yesterday, “the market may require either confirmation of a trend softening in inflation or signs that the labour market is cracking to shift regimes”.

Two questions linger. One is whether progress on inflation will get harder. Growth is still strong, the labour market is still tight, and private market rent indices have re-accelerated, suggesting shelter inflation won’t fall for ever. A second is what the Fed will do if the chances of a soft landing keep rising. Cut early to stop the lagged effects of policy from derailing the economy? Keep rates high to make sure the job is done? We’re past halftime for this rates cycle, but there’s plenty of game left to play. (Ethan Wu)

A reply on bank capital

There were quite a few lively responses to yesterday’s piece on the bank capital proposals from the Fed’s Michael Barr. Most of them were approving, but there were some interesting points of dissent. Several readers took the line that our friend Matt Klein (everyone subscribe to The Overshoot!) laid out in an email:

A partial defence of Barr: I think of capital as a way to reduce the risk of flight from uninsured depositors and other creditors, because runs are usually predicated on concerns about the distribution of losses. Silicon Valley Bank and First Republic would not have had those kinds of outflows if depositors knew that someone else would be on the hook for underwater bonds and loans . . . 

Given the outflows they had, more capital would not have helped, but maybe a different balance sheet structure would have prevented the runs in the first place. 

In other words, more capital would not have helped once the run began, but a thin capital layer made the run more likely. This is a plausible view, though I think it’s probably wrong (the Barr point I called “almost total nonsense” was the idea that different capital treatment of “available for sale” securities would have helped prevent the SVB and First Republic failures). If you read back through Unhedged’s coverage of SVB and First Republic, you might find some support for this view, in fact. We noted several times that one way to screen for weak banks is to see what capital levels would be if securities portfolios were marked to market. Here is a table from our newsletter of March 14: 

The third column, the leverage ratio, is a measure of capital strength (tier one equity capital/assets). The far-right column is the leverage ratio if capital were reduced to reflect unrealised securities losses. Back then, terms like “mark to market insolvent” were thrown around a lot, and in the case of SVB, that term was accurate. And it seems pretty safe to assume this kind of talk helped precipitate the run that overwhelmed the bank.

So you might say, with Barr, that a few more percentage points of capital might have decreased the chance of a run. Of course, First Republic was not mark-to-market insolvent — not even nearly — and it got crushed by a depositor run, too; but things are different for the second bank to fall than for the first. So there seems to be an argument here in favour of Barr’s view, which urges that higher capital levels increase “resilience”, by which he means the ability to survive losses whatever their source. 

I don’t buy this. The ultimate source of the SVB failure was catastrophically bad rate risk management combined with a flighty, concentrated, uninsured investor base. If we think that the SVB mess proves we need better regulation and supervision, the target ought to be the ultimate cause of the problem. Maybe we ought to change the way we risk weight long-term government-backed securities. Maybe we ought to have asset-liability matching rules, or require more capital just for banks with lots of uninsured deposits. Or whatever.

But to argue that all banks need more capital at all times because a tiny handful of them forgot fundamental principles of risk management seems looney to me. Because higher capital requirements have a cost. Higher capital requirements are nothing but a requirement that banks lend less, and especially when the economy is soft, we don’t want less bank lending, we want more of it. I hate to sound like a bank lobbyist, but there it is.

There may be good arguments that show that all banks need more capital. The SVB mess is not one of them.

One good read

The Economist rubbishes the concept of “greedflation”, which needed it (good week for them; they also had a nice piece on the Minivan Taliban).