By now, most regular readers know how I feel about macro aggregates: They’re largely meaningless.

Every day, policymakers of all kinds and capital allocators of all shapes and sizes consult statistical abstractions to make important (sometimes momentous) decisions with far-reaching consequences for other people’s lives and money. In some cases, those abstractions are later revised so as to bear only a vague resemblance to the “data” as originally reported.

To be sure, approximations are the best we can do when it comes to macroeconomic aggregates in large economies. Precision’s impossible. It’s thus not (necessarily) a criticism to call the data largely meaningless, as I did above. But treating the approximations as though they’re not only precise, but in fact so accurate as to make the fourth, fifth and sixth decimal places relevant for decision making, is to participate in an asinine (and potentially ruinous) parlor game. Or you might call it a casino as it relates to trading around macro releases.

With that in mind, I suppose I’m compelled to mention another “oops” moment for the BLS. Just a month after a government employee stumbled into a PR debacle by discussing methodological nuances with “super users” (a group of curious macro observers that happened to include JPMorgan and BlackRock), someone “inadvertently” published a subset of April’s CPI data to the bureau’s website half an hour early on Wednesday. The government apologized (sort of) and referred the matter to the Labor Department’s inspector general.

The mishap conjured memories of what appeared to be a leak in December of 2022, when price action in the US rates complex suggested traders got a peak at CPI data covering the prior month before it was publicly available. This time around, no one (algos included) noticed the early release, hilarious in its own right.

This is all dumber than a bag of hammers. Again: Markets and policymakers are completely beholden to statistical abstractions, and as discussed here last month, it’s increasingly apparent that traders aren’t actually wagering on the abstractions themselves, but rather on how policymakers are likely to respond to the approximations. Now we can’t even release the approximations without a scandal of some kind.

In a testament to the futility of this endeavor, note that opinions on what these approximations and abstractions are telling us about inflation vary widely, particularly once you venture outside the groupthink echo chamber.

Consider, for example, the following take on Wednesday’s CPI release as articulated by JonesTrading’s Mike O’Rourke:

The investing environment has been framed by three consecutive months of hotter-than-expected core CPI readings. In turn, the absence of a hot reading was enough to trigger a buying frenzy throughout financial markets. The three-month and six-month annualized paces [for core CPI] are both running at approximately 4%. The CPI “supercore” reading improved over March’s nosebleed reading, but the 0.421% MoM pace is a 5.17% annualized rate. The three-month and six-month annualized paces are 6.34% and 6.52%, respectively. It is hard to see much to get excited about here.

Now consider another take from SocGen’s Albert Edwards:

Measurement errors mean trying to hit [the] 2% inflation target precisely [is] poor policy (which is why a 2% target is usually interpreted as a 1-3% range). Most economists underst[and] that the disinflationary trends keeping core CPI below 2% were secular in nature and outside the Fed’s control, hence [Paul] Volcker’s point [in 2018]: Striving back then to drive CPI inflation back up to 2% via super loose monetary policy was causing distortions in the economy and speculative bubbles. The reverse argument also applies. Core CPI ex-shelter has averaged a benign 0.2% monthly rise for the last six months or so. However, beneath that calm surface, the divergence of services and goods inflation is highly unusual if not unprecedented. The Fed’s error is that in targeting one part of overall inflation (i.e., rising “supercore” services inflation), it has either inadvertently or deliberately driven core goods prices into the deepest deflation for 20 years. In fact, aside from a few months in 2003, we are now experiencing the deepest deflation in core goods prices since records began in the late 1950s. Is that sensible?

Who’s right? Because although those accounts don’t necessarily conflict, they certainly don’t have the same implications for monetary policy.

Spoiler alert: Both O’Rourke and Edwards are “right,” where that means you can make of this data pretty much anything you want to make of it.

Guess who knew this? Paul Volcker. In the 2018 article cited by Edwards (it’s actually an excerpt from “Keeping At It,” Volcker’s self-congratulatory autobiography), Volcker wrote, of inflation statistics, “No price index can capture, down to a tenth or a quarter of a percent, the real change in consumer prices.”

Wise man, that Paul. Maybe he should’ve been Fed Chair.