Seventy-five million dollars in twelve months. That is the ceiling the SEC wants to hand any qualified token issuer willing to file a disclosure package instead of a registration statement. Within weeks of that number entering the public comment record, California's Department of Financial Protection and Innovation filed a letter arguing, in substance, that the ceiling is a door and the door has no lock. The department brought receipts: nearly 200 new crypto fraud cases in its pipeline, a thirty percent year-over-year increase. Two regulators, one statute book, opposite conclusions.
I don't think this is a story about whether crypto is risky. Everyone in the room conceded that years ago. This is a story about who gets to decide — and in American securities law, the right to decide has always been worth more than any single issuance it governs.
The SEC's proposal, filed under the banner Regulation Crypto Assets, is not a novel legal instrument. It is a transplant. Read the two exemption channels side by side and the donor framework becomes visible: a Startup Exemption permitting $5 million over four years, and a Cryptocurrency Financing Exemption permitting $75 million over twelve months. The first is a compressed cousin of Regulation CF, which caps at $5 million over twelve months. The second is Regulation A+ Tier 2 with the runway squeezed from three years into one — identical ceiling, one-third the time.
That lineage tells you what the SEC is actually doing. It is not inventing crypto law. It is taking a template it has administered for a decade and re-parameterizing it for an asset class it spent four years litigating against. The pivot from enforcement to rulemaking is real, and the parameters are where the policy lives.
Underneath sits a ninety-year-old fault line. Since 1933, securities issuance in the United States has run on two layers: a federal framework plus state anti-fraud statutes and merit review. The National Securities Markets Improvement Act of 1996 preempted states from reviewing covered securities but preserved their fraud jurisdiction and their broker-dealer oversight. Every federal exemption since has been a negotiation over where that line falls. The DFPI letter is the latest move in that argument, dressed in 2026 vocabulary.
The timing is not accidental. MiCA is phasing into full effect across the EU under an authorization model, where an issuer seeks approval before it touches a public market rather than after. Tokenized treasuries have moved from proof-of-concept to allocation. ETF wrappers have normalized institutional exposure. Capital is actively comparing jurisdictions in a way it was not doing in 2023, and a federal exemption framework is the United States bidding for a share of that flow. Which is precisely why a state regulator with its own enforcement statistics feels its leverage slipping.
The comment record is unusually well-populated. Ripple, Coinbase, and Nasdaq all filed in support of adjustment. That alignment is the first thing worth interrogating, and it comes back later.
Start with the economics, because the economics are where the narrative collapses.
$5 million over four years is $1.25 million a year. That is not a token launch. That is a seed round with extra paperwork. The Startup Exemption is calibrated for pre-product teams, and any project treating it as a fundraising strategy is announcing, in a public filing, that it has no institutional demand. The interesting number is the second channel.
$75 million in twelve months against Reg A+ Tier 2's $75 million in thirty-six months. The SEC did not raise the ceiling; it tripled the velocity. For a mid-cap protocol that has already run a private round and wants to avoid a full registration statement, that compression is the entire value proposition. A three-year Reg A+ qualification — audited financials, state notice filings, ongoing reporting — is a commitment most crypto treasuries will not make. Twelve months is a sprint they can staff.
Now the part the press releases skip: the compliance cost curve is fixed, not proportional.
Disclosure-based regimes are cheap at scale and brutal at the bottom. A qualified audit, a securities lawyer on retainer, a finance function that can produce semi-annual statements, a transfer agent, an investor-relations obligation — that stack runs somewhere between $400,000 and $1.2 million a year depending on jurisdiction and complexity. Against a $75 million raise it is noise. Against a $5 million raise amortized across four years it is a double-digit percentage of gross proceeds before a single token moves. The Startup Exemption's headline generosity evaporates on contact with the fixed cost of being a reporting issuer.
I have watched that arithmetic kill deals. In 2025, advising three token issuers on MiCA positioning ahead of the EU's phased deadlines, the conversation never opened with narrative. It opened with compliance burn rate — what an authorized-entity structure costs per quarter, and whether the treasury survives eighteen months of it before revenue arrives. Two of the three restructured. One deferred its European launch entirely and moved to a jurisdiction with a lighter authorization path. None of them made that call because the rules were unclear. They made it because the rules were clear and the number did not work.
That is the lesson the DFPI letter circles without naming. Clarity is not viability, and a permissive exemption carrying an expensive reporting tail is a trap dressed as an opening.
Which brings us to the mechanism that actually decides the outcome. The proposal preserves state authority over securities fraud, deceptive practices, and unregistered broker-dealer activity. Read that list again. Every item is post-hoc. Fraud investigation happens after the money is gone. Broker-dealer enforcement happens after the sales have cleared. What the federal exemption removes is the state's pre-emptive gate — the merit review, the qualification filing, the ability to say no before an offering goes live.
DFPI is not confused about what it is losing. It is naming it precisely. The thirty percent fraud increase is not an argument about crypto's riskiness in the abstract; it is an argument about the only lever the department still holds if the exemption passes. When your gatekeeping authority is preempted, your enforcement statistics become your entire political identity. Expect those numbers to be published quarterly and cited often.
The Howey question everyone keeps asking is the wrong question. An exemption is not a classification. The proposal does not declare tokens non-securities. It says: these remain securities, here is a lighter path to offering them, and here is the disclosure you owe while you do. Money in, common enterprise, expectation of profit, efforts of others — nothing in a $75 million exemption changes how a court reads those four prongs. What changes is the price of complying with the answer.
The institutional read matters more than the retail read here, and my own work colors this. In 2024, pitching tokenized treasuries to Auckland-based funds, I built a proof-of-concept dashboard with three developers to answer the question allocators kept asking. Not what is the yield. What happens if the issuer stops reporting. The dashboard was never a product; it was an argument about disclosure continuity. The funds that engaged signed because they could see a reporting cadence, not because they loved the asset class. That instinct is exactly what the ongoing-reporting provisions are designed to satisfy, and it is why the details the comment letters have not yet specified — frequency, format, audit standard — are the real battleground. Everything else is positioning.
There is a second-order effect worth pricing. If the final rule creates a federal exemption path while states retain only enforcement authority, the United States drifts toward a two-track compliance market: issuers satisfying SEC conditions and, simultaneously, whatever broker-dealer licensing and anti-fraud exposure their state of operations imposes. Nobody designs that outcome. It emerges. And the beneficiaries are predictable — law firms, audit shops, compliance software vendors, and the advisory layer that translates one regime into the other.
Here is the part that runs against how this letter is being read.
I don't buy the framing that DFPI's objection threatens the proposal's survival. A single state agency's comment letter does not move a commission vote. The final text will be shaped by the SEC's internal coalition, by congressional pressure, and by ninety days of lobbying. California's letter is one input, and not the heaviest.
The contrarian read is that the industry should be quietly grateful the letter exists. A final rule that gutted ongoing reporting to placate state regulators would produce an exemption no institutional allocator would touch — a $75 million channel that reads, to any fund's diligence team, as the lighter-disclosure bucket. The DFPI objection gives the SEC political cover to keep the audit and reporting teeth in the final text while telling critics it listened. State pushback is the mechanism by which the exemption becomes credible, not the mechanism by which it dies.
Which is also why I don't read Ripple's, Coinbase's, or Nasdaq's signatures as unconditional endorsements. Support for adjustment is a position compatible with almost any final text. The gap between supporting a framework and accepting a specific version of it is where every compliance budget gets made or broken. Nasdaq's participation is the tell. A listing infrastructure operator entering a token issuance debate is not lobbying for retail access. It is positioning for a tokenized-securities venue, and it wants the disclosure standard set at a level its existing surveillance and listing machinery can absorb.
The comment window runs its usual ninety to one hundred twenty days. Watch three signals: whether NYDFS or the Massachusetts Securities Division files alongside California, converting a letter into a bloc; whether the audit and reporting requirements survive in the final text; and whether any supportive commenter re-files once it sees actual terms. This proposal will not move a spot price. It will move the cost of raising capital in the United States for a decade — and the difference between a $75 million exemption that works and one that does not is a paragraph about reporting frequency almost nobody is reading.