Inflation, still scary

Receive free US updates

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

Good morning. Ethan here; Rob returns tomorrow. Is all the good news priced in already? So far this earnings season, the median price reaction to an earnings beat has looked limp versus history, according to Morgan Stanley. One notable exception: Berkshire Hathaway, which rose 3.4 per cent after Saturday’s earnings, and after Unhedged wrote that Berkshire probably can’t sustainably beat the S&P 500. Readers can decide which seems the likelier catalyst. Email me: ethan.wu@ft.com.

Base effects and the inflation threat

Here is a dumb point about inflation “base effects” that’s nonetheless easy to mix up. Inflation is a rate of change, so the denominator you choose to measure it with makes a big difference. The two standard methods are month-over-month and year-over-year comparisons. What’s tricky is that you can have flat monthly rates alongside rising annual rates. Whatever happened to prices a year ago ripples over to today.

Though this point is fiddly, it matters. Because the sharpest inflation increases happened in the first half of 2022, annual comparisons for the rest of this year’s CPI reports will be boosted by more slowly rising denominators. The dominant soft-landing market narrative is anchored in part by the pretty little 3-handle on current headline annual inflation. If this starts rising, even if it reflects a tame monthly rate, investors could begin doubting their assumptions about how immaculate disinflation will be.

This basic point is made in a recent post by Rob Arnott and Omid Shakernia of Research Affiliates. Earlier in the year, they forecast “an illusion of tumbling inflation in the first half of the year, perhaps even falling below 3 per cent by mid-year, then an illusion of soaring inflation in the second half of the year”, driven by shifting base effects. They illustrate why in the chart below, first published in January. The orange line shows how high annual inflation could conceivably rise, assuming flat monthly CPI readings at 46 basis points (the 2020-23 average). The green line shows the lower bound, with monthly readings at zero throughout 2023:

A 46bp monthly print is unlikely, as Arnott and Shakernia concede. (The 2023 monthly average so far is 27bp.) Even still, they think year-over-year headline inflation could finish the year in the range of 4-5 per cent, starting with Thursday’s July CPI report. From their post:

Suppose, for example, that we see an average of 20-40 basis points of inflation each month through year-end. Then we should see the year-over-year inflation rise to somewhere between 4% and 5½% by year-end. That’s enough to be alarming to most observers, who are typically not paying attention to the low rates of inflation that we are replacing late 2022.

We find it a bit amusing that, if inflation finishes the year at half its peak levels of mid-2022, this will be an adverse shock to many (indeed, likely most) investors. If we finish the year at 4½% inflation, we will be pleased by this usually benign denouement following a dangerous inflationary surge, but most of the investment community, media, and political elite will likely be alarmed (fear sells!).

Research Affiliates is not the only shop thinking about this. Bank of America’s chart below received a good deal of attention back in July. It shows where yearly inflation would go assuming a few different flat monthly rates (the chart is misleading; more on that below):

Omair Sharif of Inflation Insights has a beef with BofA. Their chart’s title suggests that any monthly inflation reading above 20bp would push up annual headline rates. That would seem a low bar, raising fears of a base effects-driven inflation resurgence. But Sharif points out that BofA misleadingly uses non-seasonally adjusted monthly rates. If you use more conventional seasonally adjusted numbers instead, base effects look much less threatening. The real bar for monthly rates creating inflation re-acceleration is more like 40bp. Here is Sharif’s redo of BofA’s chart:

Base effects should be easy enough to ignore in the next few months because, as Carl Riccadonna, chief US economist at BNP Paribas, pointed out to me yesterday, “when we talk about base effects we’re basically talking about energy base effects”. Watching core inflation, even in year-over-year terms (currently at 4.9 per cent), will probably give you a better idea of where inflation is heading. But in general, “for the next couple months, you shouldn’t be too focused on the year on year trend”, he added.

So, all good then? Not entirely. Rather than a sharp inflation rebound, the problem is if inflation settles in at an uncomfortably high rate. Over the next couple months, we will probably see encouraging monthly core inflation readings around 20bp, thanks to disinflation in shelter and used cars. But Sharif thinks the next 12 months will look more like core inflation pinging between 20bp and 30bp, somewhere between the high-2s and 4 per cent in annualised terms. Over at The Overshoot, Matt Klein makes the case that it’ll look more like 4:

Without a downturn, it is unclear whether (or why) wage growth would decelerate enough from here to bring inflation all the way back to the Fed’s 2% yearly goal over any reasonable timeframe . . . 

The main reason to worry is that nominal wages have consistently been rising about 5%-6% a year since last summer . . . Since consumer spending tracks wage growth better than anything else, real volumes of goods and services would have to rise about 3%-4% a year for the current pace of wage increases to be consistent with 2% inflation. That would certainly be my preference, and there are good reasons to think that productivity might accelerate, but the likelier outcome is that underlying inflation is closer to 4% than 2%.

This is the hole in the soft landing story: persistently high wage growth sustaining consumer spending. In lieu of slower wage growth, falling prices may just increase consumer purchasing power, lifting demand in a way that, ultimately, shows up as higher prices.

Sharif gives a vivid example from the used-car market, in principle one of the most rate-sensitive parts of the economy. In the last six months of 2022, used-car prices fell steadily. Then, in January, demand popped. Driven by lower prices (and substitution from new cars), retail used-car sales shot up 10 per cent in just one month. The surge in demand “caught all the dealers off guard”, forcing them to rush to wholesale auctions, says Sharif. Three months later, that mad dash showed up in inflation. The used cars and trucks CPI sub-index rose more than 4 per cent in April and May.

Used cars are an especially dramatic category, and probably unrepresentative. But the point is that inflation can only fall as far as consumer spending lets it. The inflation threat to markets has not gone away.

One good read

A very nice book review considers the gap between George Orwell’s rhetoric and his life.

FT Unhedged podcast

Can’t get enough of Unhedged? Listen to our new podcast, hosted by Ethan Wu and Katie Martin, for a 15-minute dive into the latest markets news and financial headlines, twice a week. Catch up on past editions of the newsletter here.