There were two main takeaways from Jerome Powell’s press conference on Wednesday, when the Fed lowered its growth outlook and raised its inflation projections while keeping rates on hold.

The first was that progress towards the FOMC’s inflation target has stalled, and that as such, the Committee’s arrival at 2% will be delayed, in part due to tariffs.

The second was that it’s too soon to say how Donald Trump’s policies will impact the macro picture in the US.

There’s some tension there. If the Fed now thinks it’ll take longer to cajole underlying price growth back down to 2% sustainably, and if the reason for the delay is tariffs, then it’s not to soon to say how Trump’s policies will affect the macro outlook, because… well, because you just said it.

The key question, then, is whether any tariff-related price shocks are a one-time thing. Whether they’re — don’t laugh — “transitory.” “How much of the higher inflation forecasts is due to tariffs?” Howard Schneider, from Reuters, wondered, referencing the higher PCE projections in the new SEP. “Is this a one-time level price shock?”

Powell stated the obvious. It’s “very difficult” to develop a “precise” assessment of how much of any given inflation impulse is solely attributable to tariffs. “The answer is clearly some of it. A good part of it is coming from tariffs,” he said, adding that the Fed will try its best to separate “tariff inflation from non-tariff inflation.” For now, the “base case” is that tariffs are probably a one-off price level increase.

Colby Smith, now at The New York Times, asked about the jump in inflation expectations. She didn’t specify the enormous two-month surge on the University of Michigan’s five- to 10-year measure in February and March, but thankfully, another reporter did later in the proceedings.

“We do monitor inflation expectations very, very carefully,” Powell told Smith, on the way to asserting that although near-term inflation expectations might’ve risen broadly, longer-term expectations are anchored, as are market-based measures. I don’t know about anyone else, but the figure below doesn’t look “anchored” to me.

That is, of course, the University of Michigan gauge mentioned above, and it looks suspiciously (dangerously) like unmooring.

Every so often, including late last week, I retell the story of the Fed’s first 75bps rate hike in 2022 at that year’s June FOMC meeting. My narrative says it wasn’t the May 2022 CPI report which sealed the deal on 75bps increments, but rather an overshoot on University of Michigan longer run inflation expectations in the preliminary read for June. Again, I repeat that narrative every time that series overshoots, and the preliminary reading for this month was the biggest overshoot in decades.

Powell on Wednesday basically said that version of the June 2022 rate hike story is fictionalized. When asked specifically whether he no longer cares as much about that measure (the University of Michigan series) as he did three years ago, Powell said, “I mentioned it back then [but] in no way did I place huge weight on it.”

Folks, that’s as close to a lie as I’ve heard Powell tell. I’m not going to go back and parse Fed communications from that period, but I don’t misremember details like this. This isn’t a game of telephone where the story gets embellished every time it gets retold. Powell and the Fed were unnerved in June of 2022 by a high print on that measure of household inflation expectations, and that high print fed into the decision to go with upsized, 75bps rate hikes. Full stop.

I like Powell, really I do, but he’s done this before, which is to say he puts a lot of emphasis on this or that metric when it serves the Fed’s purposes, only to turn around and deemphasize the very same metric later, when it doesn’t. For a lawyer, he’s not very good at hiding that penchant for pseudo-dishonesty. In fact, he’s a terrible liar in general, which makes me wonder how he could’ve been a lawyer in the first place, but that’s another discussion.

Bottom line: Longer run inflation expectations came unglued in the Michigan poll this month and last. Any Fed official who suggests that poll doesn’t carry more weight than a majority of other such polls, isn’t telling you the truth. I guess if the Employment Cost Index — which Powell placed a ton of emphasis on previously — were to accelerate again, he’d pull a Trump-with-the-accusers act vis-à-vis the ECI: “I never ‘moved on’ this index. I’ve never even seen it before!”

Smith went on to ask about consumer confidence and the extent to which a decline in household sentiment might ultimately translate into weaker spending and so on. Powell said that for now, the hard data’s still solid. Inflation “has started to move up, partly in response to tariffs” and some of the soft data does point to a slowdown, Powell conceded, but “the relationship between survey data and actual economic data hasn’t been very tight.”

Asked by Wall Street Journal “Fed whisperer” Nick Timiraos whether inflation in the US as it stands today meets Alan Greenspan’s definition of price stability (i.e., Expected changes in the general price level do not effectively alter business and household decisions), Powell obfuscated, but ultimately said no, without using that word. “I think we were getting closer and closer to that,” he said. “With the arrival of tariff inflation, further progress may be delayed.”

Pressed on whether the Fed’s independence is in jeopardy now that Trump’s taken a shine to firing people he arguably isn’t allowed to fire, Powell referred to his answer from the November FOMC, when he said, “Not permitted under the law.” The reporter scoffed. Literally. He laughed out loud, as if to say, “Mr. Chairman, have you not been paying attention to what’s going on?”

Finally, one reporter wondered whether it’s “hard to get clarity in a government-by-tweet?” Powell hesitated, then stammered: “These decisions are going to be made and they’re going to be implemented and we’ll be adapting as we go.”