The Powell Fed’s in a tough spot.
That’s a good early candidate for macro-market understatement of the year.
Any economist will tell you that consumer inflation expectations are a key component of the price formation process. That’s “canonical,” as the Cleveland Fed put it a few years back.
The Fed was free to fret openly about de-anchored household inflation expectations during 2022 because back then, the culprit was an external twin supply shock: The pandemic and, later, the war in Ukraine.
Yes, domestic fiscal stimulus played a part in driving up prices, and higher prices can be a self-fulfilling prophecy through the expectations channel. But justifying a hawkish policy bent by reference to inflation expectations wasn’t necessarily tantamount to a critique of government policy, unless by “government policy” you meant the Kremlin’s “policy” of reconstituting an empire in Eastern Europe.
In 2025, household inflation expectations on the most widely-cited measures are out of control, where that means perched at multi-decade highs after inflecting on a 90-degree angle over the past three months. This time around, the Fed can’t put as much emphasis on those metrics because to do so is to implicitly critique US government policies, specifically the Trump administration’s determination to deport millions of low-wage workers and, more poignantly in the context of near-term inflation, a 10-fold increase it the average US tariff rate.
It’s not that the Fed’s shied away completely from discussing the macro implications of tariffs — Powell was pretty open about the extent to which a trade war threatens to push the Fed away from its goals in his last public speaking engagement — it’s that officials have expressed almost no alarm whatever about the historic spike on the University of Michigan’s inflation expectations series, and I doubt that’s because they aren’t worried.
Powell and his colleagues were palpably — publicly — concerned when those same measures exhibited what, by comparison, was a tame increase in 2022. If they were to scale up that worry and the attendant public hand-wringing commensurate with the relative size of the spike on the offending metrics in 2025, they’d be openly discussing rate hikes right now, not debating when the next cut might arrive. The reason they aren’t is fairly obvious: Trump’s got it in for Powell and for the Fed in general.
Have a look at the simple chart below, which just plots the Michigan five- to 10-year series (that’s the important one) with YoY headline CPI.
Suffice to say there’s no precedent for a meaningful uptick on that measure in the absence of a concurrent increase in realized inflation, let alone a monumental spike like that witnessed over the last three months.
Simply put: Were it not for the political sensitivity of this issue, Powell and co. wouldn’t be anywhere near as sanguine about inflation expectations as they currently come across in public speaking engagements.
Why does this matter beyond the obvious — i.e., beyond the fact that Americans are terrified of “Tariff Man” and so is the Fed? Well, because if inflation expectations are a key input in the price formation process, then a sustained increase in those expectations should be expected to exert upward pressure on prices as consumers pull forward purchases. More simply: An inflation expectations panic can quickly self-fulfill.
That’s the macro side. On the markets side, elevated inflation has implications for the stock-bond return correlation, which tends to flip positive in the presence of consumer price pressures. As we saw in 2022, that’s very bad for balanced portfolios, by definition.
“Inflation dynamics are a major driver of the correlation between stock and bond returns,” JPMorgan’s Thomas Salopek wrote, in a new QDS piece. “During periods of low inflation, the correlation is typically negative [while] during inflationary periods, the stock-bond correlation usually becomes positive.”
The figure on the left, above, shows the historical relationship between core CPI and the rolling 12-month forward stock-bond correlation. The takeaway: When inflation creeps up beyond 3.5% or so, the stock-bond correlation typically flips.
In the same note, Salopek looked at what the implications of such correlation flips are for multi-asset portfolios with a target vol constraint. Not surprisingly, negative stock-bond correlations (i.e., diversification regimes) are associated with lower vol and vice versa. To channel Charlie McElligott, vol’s the “exposure toggle” in modern markets. What’s another word for “exposure”? Leverage.
The figure on the right, above, illustrates the point. “When the [stock-bond] correlation is strongly negative the portfolio benefits from lower volatility [and] will lever up,” Salopek went on. “When correlation is positive, diversification is reduced and the portfolio volatility is higher, requiring less leverage to meet the target vol.”
Coming full circle, if the Fed’s constrained not only in its capacity to address soaring household inflation expectations with policy, but even in its capacity to address the issue in public communications, those expectations could become embedded in consumer psychology leading to i) higher realized inflation, ii) a more persistently positive stock-bond correlation and iii) a de facto cap on portfolio leverage in the presence of elevated volatility.
The Fed probably sees high risk of both inflation and weak employment. Both may develop imminently and the Fed may be forced to choose between its mandates, while responding to market stress.
Trump is trying to set the Fed up as the fall guy. He won’t convince anyone but diehard MAGAs that the Fed is to blame for tariff-driven inflation, but if the Fed raises rates into a developing recession he may have a second chance to deflect blame.
Tariff driven inflation may be beyond the Fed’s ability to tame anyway. If a product comes from China, is tariffed by 50% to 145%, and in the medium term sourcing a substitute from elsewhere will cost 2-3X the China cost, then the price is going up by say 20% to 40% depending on required margin structure, rather independent of whether demand weakens.
And all this could well be “transitory” – not in the economic sense, but in the political sense. Companies and investors get whipsawed and c’est la vie, the Fed gets whipsawed and you’ve got the next Arthur Burns.
Given that, I don’t think it would make sense for the Fed to raise rates in response to rising inflation expectations, and maybe not even actual rising inflation before seeing how employment is responding.
Supply chain lags is probably about 1.5-2 months for shipped shipped by ocean from China, most of which is not electronics temporarily exempted. Some US companies built up inventories in 1Q, smaller ones had less ability to do so. Hence WMT and others warning that shelves may start looking bare in a month or two. Prices are rising now for some things but the broader inflation will show up in the coming couple months.
Employment weakness probably starts showing up in openings and hires over a similar period, with some industries – transportation, etc – leading.
Fed may want to see how that develops before deciding on next steps.
Prophylactic inventory buildups will alter profit performance, depending of how large they are and how long they last. If consumers slow down consumption while inventory builds up for too long profits could well be dented.
Yes, high inventory is usually a leading indicator for profit pressure. In this case . . . not sure. Uncharted waters we sail.
I think that bear needs to be available with the mugs and t-shirts.
The AI image generation is getting so crazy. I showed it my old Twitter banner image today — which was pretty impressive on its own to have come from a robot — and asked it to generate an improved version, and this is what it came up with: https://heisenbergreport.com/wp-content/uploads/2025/04/TwitterBanner22025New_LE_upscale_balanced_x4-scaled.jpg
It took less than five seconds for it to analyze the old banner and produce that new one and notably, I couldn’t critique it. There’s nothing to improve on. And yet, somehow, in a month or two, some new version of the same AI will find a way to make it even better.
Which AI image generator did you use there? Agreed, this stuff is already impressive, and improving rapidly too.
BEI is still anchored
When was the last time you heard a family of four on $75k/year say, “Well, we were worried about inflation, but then we checked our Bloomie and as it turns out, there’s nothing to worry about.”?
Put another way — and somewhat ironically — anyone who knows what BEI stands for doesn’t count in the price formation process, because if you know what BEI stands for, you probably make enough money that you don’t base your consumption decisions on inflation expectations.
I think long before we see real tariff-driven inflation, we are going to need to see real imports and real tariff assessments. Shipping and freight is already gunked up as ports get clogged with empty containers going nowhere, while trade zones and temp warehouses fill up with stuff not willing to be imported without tariff clarity. Lots of truckers getting dead headed on the West Coast dropping stuff off with nothing to bring back. Guessing that less than full load sea and land transportation costs won’t help out on the inflation front, nor the national hit to productivity from just trying to follow the plot line of this circus.
People are throwing the term around more and more, but I doubt Trump will satisfy himself with some run of the mill stagflation. That may actually turn out to be wishful thinking as the world’s greatest dealmaker flirts with Third Worldian slumpflation, coming soon to an urban center near you.