economy Markets

Credit Suisse: The Death Of The Phillips Curve Has Been Greatly Exaggerated

"The market appears to have overlearned the lesson of the past few years."

Over the past 12 or so months, we’ve spilled gallons upon gallons of digital ink in these not-at-all-hallowed pages expounding on the Phillips curve.

A couple of years back, this normally wonkish topic crawled out of your dusty Econ textbook, took an axe to the doors of mainstream financial media outlets (Jack Torrance style) and became an unlikely headline-grabber as policymakers the world over struggled to explain an apparent breakdown in one of economics’ most fundamental relationships.

Despite becoming a fixture in mainstream financial media articles about post-crisis monetary policy, the Phillips curve still lacks the kind of pizazz that makes for compelling reading. Fortunately, Deutsche Bank’s Aleksandar Kocic has the antidote if you’re suffering from bland Phillips curve commentary. We’ve employed his brilliantly colorful characterization before and we’ll do it again here. To wit, from a June note:

Through the last four cycles, Phillips curve has asserted its importance in an unorthodox way and, as such, attained a special status; it inhabits a space different from other macroeconomic frameworks and metrics. In each cycle, it falls apart, but after every annihilation, it re-composes itself and continues to play an important role. It appears “indestructible”, but not in a conventional way, more like a survivor of one’s own death. Phillips curve functions like an organ without a body, an equivalent of Cheshire cat’s smile (in Alice in Wonderland) that persists alone, even when the cat’s body is no longer present. This time is no different in our view. The figure shows the Philips curve through several cycles starting in mid-1980s. Each cycle has a different color which implicitly marks their beginning and end. The first thing one observes is that this is more of a “spaghetti” then a curve. The main reason is that it captures different cycles – a testimony to its falling apart and recomposing itself after each.


He’s updated that analysis a number of times since that was written, and we’ve referred to that passage on countless occasions in the context of, to name a few topics, Donald Trump’s fiscal policies, US trade policy and Jerome Powell’s October comments regarding the prospect that the Phillips curve might one day seek its “revenge.”

Read the Cheshire Cat series

Cheshire Cat’s Smile: ‘What Is The Fed Worried About?’

The Real Estate Developer, The Chinese Strongman And Cheshire Cat’s Smile

Cheshire Cat’s Revenge? Jerome Powell Says Phillips Curve Probably Won’t Kill Us All In Our Sleep

30° Of Separation And Cheshire Cat’s Revenge

This discussion invariably comes up whenever folks start to talk about the Fed using still-subdued inflation to justify “patience”. Obviously, that is the path the committee is on currently and more than a few commentators have suggested that a “rogue” inflation print (or two or three) is a risk to the extent it could force the market to start pricing back in the possibility of rate hikes. The February jobs report was interesting in that regard, as AHE printed a new cycle high.

It’s with all of that as the backdrop that Credit Suisse’s James Sweeney is out with a new note called “The death of the Phillips curve has been greatly exaggerated.”

Again, this never makes for the most compelling reading, but it’s an important topic and as such, we thought we’d highlight a few passages and visuals from the note, dated Monday.

After noting that inflation hasn’t been particularly responsive to the labor market in recent years (see visual below), Credit Suisse describes the state of the debate as follows:

This has prompted some soul-searching among inflation forecasters, both in markets and at the Fed. Some insist that the Phillips curve has become “flat,” meaning it takes large swings in unemployment to have a discernible impact on inflation. A recent speech by Fed Chair Powell lays out the case for a flatter Phillips curve, noting that anchored inflation expectations may account for the changing relationship. Some have gone farther and concluded that the Phillips curve relationship has broken down entirely. For instance, in the January 2018 FOMC minutes, some participants “questioned the usefulness of a Phillips curve-type framework.”


For Credit Suisse, this isn’t as “puzzling” as it’s made out to be.

After stating the obvious (i.e., that AHE and ECI have accelerated to post-crisis highs), the bank notes that “the most labor-sensitive parts of core PCE have responded by breaking higher” too. Specifically, Credit Suisse writes that if you strip out “shelter, medical services, and financial services [which] have their own idiosyncratic drivers, other service prices are well correlated to wage costs.”

In fact, “other services” are sitting at post-crisis highs, and if you ask Credit Suisse, will likely rise even further in the months ahead.


(Credit Suisse)

Sweeney goes on to warn (if that’s the right word here) that “idiosyncratic headwinds” to other PCE components may well abate.

“Importantly, there is no reason to expect these headwinds to persist indefinitely [as] stabilization in the housing market and a pickup in global growth could see shelter and goods prices reaccelerating by 2020”, he goes on to say, adding that “health care inflation could also see a shift higher next year as legislative and regulatory headwinds fade.”

The overarching message from Credit Suisse is that while inflation may be subdued in the near term, that could change. Here’s the money quote from Sweeney:

The market appears to have overlearned the lesson of the past few years and doubts that a tight labor market can push inflation sustainably higher. Looking at the details closely though, the recent benign behavior of inflation looks less like a fundamental rule and more like a convenient coincidence. We do not expect inflation to rise in the near term, but it has already begun percolating beneath the surface. If a few more things go right, core PCE could easily be above target and increasing by next year.

It’s a good thing Powell and co. are willing to tolerate an overshoot. Otherwise, the market might be forced to start doubting whether this hiking cycle is in fact over. We’re going to go out on a limb here and say that risk assets would not like it if STIR traders started pricing in hikes again in the event price pressures start to materialize in earnest…


1 comment on “Credit Suisse: The Death Of The Phillips Curve Has Been Greatly Exaggerated

  1. Phillips Curve, CAPM, EMH, Natural Rate of Unemployment, Neutral Rate. All useful academic concepts. All useless to apply in a policy making framework. If the Phillips Curve is unstable over time, what possible use is it for a policymaker? Not much. Same as the neutral rate of interest. The problem is not the theory it is the misguided application of it.

Speak On It

Skip to toolbar