Most writing about earned wage access regulation answers "which states have a law." That is a solved question and a dozen firms answer it well. The question nobody answers is which legislatures are moving right now — and the honest version of the first question, which is how would you know. This post gives you both, including the part where our own numbers are lower than the ones you have read elsewhere.
Nine states, twelve acts — each with an enactment action in the record
As of the pull on 1 September 2026, EWA Watch holds twelve enacted acts across nine states, each one a bill whose title names earned wage access and whose legislative record carries a formal enactment action:
| State | Act | Enacted |
|---|---|---|
| Arkansas | HB 1517 — Earned Wage Access Services Act | 20 Mar 2025 |
| Indiana | HB 1125 — Earned wage access services | 6 May 2025 |
| Kansas | HB 2591 — EWA registrants brought under the financial institutions information security act | 9 Apr 2026 |
| Louisiana | HB 368 — earned wage access services | 1 Jul 2025 |
| Maryland | HB 1294 — Credit Regulation & EWA Modernization | 25 May 2025 |
| Maryland | SB 94 — Earned Wage Access Revisions | 28 Apr 2026 |
| Nevada | SB 290 — employer-integrated EWA regulation | 13 Jun 2023 |
| South Carolina | S 700 — Earned Wage Access Services Act | 21 May 2024 |
| Utah | HB 279 — Earned Wage Access Services Act | 26 Mar 2025 |
| Wisconsin | AB 574 — regulating EWA services | 21 Mar 2024 |
| Wisconsin | AB 1025 — earned wage access services | 3 Apr 2026 |
| Wisconsin | AB 930 — cross-reference fix, DFI remedial | 27 Mar 2026 |
Two things worth noticing. Maryland and Wisconsin appear more than once — states amend their EWA statutes, and the amendment is the version that binds you. And Kansas got there through an omnibus banking bill whose title runs to four sentences, one of which is about EWA registrants. A list of "states with EWA laws" will never show you that.
The three we cannot confirm — California, Connecticut, Missouri
Published inventories put the number at twelve. We show nine, and we would rather explain the gap than quietly print somebody else's number as though we had verified it.
- California. Our pull returns exactly one California bill, and it is about the Political Reform Act and tribal-government compensation. California's EWA rules did not arrive through a bill whose text our queries reach. We are blind here and we say so.
- Connecticut. We hold two Connecticut EWA bills — HB 5007 and HB 5140 — but neither carries an enactment action. Connecticut's 2026 act is not in the record we pull.
- Missouri. Six Missouri bills, five of them title-confirmed EWA bills, none showing enactment. The Missouri vehicle appears to be an omnibus — SB 103, titled "Modifies provisions relating to judicial proceedings," signed 6 July 2023 — which our classifier holds in the flagged tier because its title does not name EWA. It is in your brief, marked, for a human to judge.
If you operate in those three states, use a law firm's inventory, not ours, for the enacted question. We will tell you that rather than let you find out.
How we decide a bill became law, and why we changed it
Until yesterday our tracker decided enactment by looking for words like "signed," "enacted," "approved by the Governor" or "chapter" in the latest recorded action. It sounds reasonable. It is wrong, because legislatures do not agree on the words:
- Arkansas HB 1517: "Notification that HB1517 is now Act 347"
- Wisconsin AB 1025: "Published 4-4-2026" — in Wisconsin, publication is the final step
- Louisiana HB 368: "Becomes Act No. 496 without the Governor's signature"
Three enacted statutes; none of them matches the keyword test. So a Wisconsin provider's brief would have read "no enacted EWA law found in the states you operate in" while three Wisconsin Acts were on the books.
We replaced the guess with the record. Enactment now comes from the legislative record's own structured action classification — the same field a legislature uses to mark a governor's signature or a bill becoming law without one — and it is stored, not re-inferred each time someone runs a report. The change found three states we had been missing and added no false ones: every state it now reports appears on the published inventories too.
We are telling you about a defect in our own product because the alternative is that you trust the number without knowing what produced it. That is the whole business.
What is moving: 96 bills across 28 states
The enacted list is the smaller half. From the same pull:
- 171 bills matched across all states and sessions.
- 122 name earned wage access or a synonym in the bill's own title, across 37 states.
- 96 of those are still live — not enacted, not dead — across 28 states.
- 37 have had action in 2026, across 14 states: Michigan (12), Wisconsin (4), Kansas, Minnesota and New Jersey (3 each), Maryland, New York and Washington (2 each), then Arizona, Colorado, Georgia, Maine, Massachusetts and Oklahoma.
- 49 matched on full text only, across 15 states — the flagged tier below.
Michigan is running a twelve-bill package
The largest single concentration of 2026 activity is a state nobody lists as an EWA jurisdiction. Michigan has twelve title-confirmed bills, HB 5558 through HB 5569, all of which moved on 30 April 2026 — HB 5558 with its committee recommendation concurred in, the other eleven referred to second reading together.
They are not twelve drafts of one bill. Read as a set they amend money transmitter provisions, wage and fringe benefit regulation, collection practices, the definition of a financial licensee, deferred presentment service transactions, the credit reform act, the regulatory loan act and the criminal usury statute — with the licensing of earned wage access services running through them.
That is a state deciding EWA's regulatory category rather than bolting a fee cap onto the existing one. If you operate in Michigan, the month it sits at second reading is a cheaper month to think about it than the month a registration window opens.
Why a keyword search over-reports, and what we do instead
A full-text search of state legislation for "earned wage access" returns bills with nothing to do with it. Real examples from our own data:
- Utah SB 73 — Online Age Verification Amendments
- Utah HB 72 — Criminal Use of Cryptocurrency Amendments
- Oregon HB 2930 — Relating to conflict of interest requirements
- Texas HB 14 — support for the development of the nuclear industry
All four are enacted laws. None of them regulates EWA. Counting them would inflate your compliance surface with obligations that do not exist — and note that a naive "enacted bills mentioning EWA" count would have handed you twelve states including Texas and Oregon, which is how a wrong number can look like the right one.
Dropping them silently is worse, because some are real: New York A 258 is "Provides for income access services in the state", Nevada SB 44 is "Revises provisions relating to financial services", and Missouri SB 103 above. No title names EWA. All are worth a look.
So we do neither. Every bill is classified and the classification is stored — title match, or full-text match only — and you see both, labelled, with the tail in its own section. 49 bills across 15 states currently sit in that flagged tier. We will not assert they regulate you, and we will not hide them from you.
The federal bill does not settle this
H.R. 9330, the Earned Wage Access Consumer Protection Act, was introduced by Rep. Bryan Steil in June 2026 and advanced out of the House Financial Services Committee at the end of that month. As reported, it would set a federal framework and would not classify EWA as credit. Analyses: Consumer Finance Monitor, Faegre Drinker.
A note on that paragraph, in the same spirit as the rest: two reputable sources give different vote tallies and dates for that committee action, and we could not reach the primary record to settle it. So we printed no tally. If a number is not checkable, it does not go in.
A committee vote is not a law. Until it is one, the states keep legislating.
If you operate in more than one state
- Separate enacted law from moving bills. One is an obligation with a date; the other is a decision about whether to keep operating there at all.
- Track the states you are actually in. A fifty-state brief is mostly noise to a provider live in four, and noise is what stops people reading the brief.
- Insist on a link to the bill record for every line, and check whether it points at the legislature or at somebody's summary. If a claim cannot be traced, you will end up trusting it on exactly the day it goes stale.
What EWA Watch is, and what it is not
EWA Watch is that tracked record as a product. You tell us which states you operate in; the brief covers those, ordered the way you would act on it — enacted law first, then bills still moving, then bills that died this session so you stop chasing them — with the flagged tier kept separate and marked. Every one of the 171 tracked bills currently links to the legislature's own bill page: arkleg.state.ar.us, le.utah.gov, docs.legis.wisconsin.gov, kslegislature.gov. Where a state does not publish one, we link a bill-tracking aggregator and label it as one. The record is re-pulled daily. $299/month up to five states, $999/month for all fifty.
What it is not, plainly: it is not legal advice and not a compliance determination. It does not interpret statutes, does not complete or file licence applications, and does not certify that you are or are not compliant. It will not even give you an effective date for an enacted act — that lives in the enrolled text, and we will tell you to read it there rather than guess on your behalf.
The same discipline, again
We keep arriving at one question in unrelated domains: what is this number actually grounded in? In patent due diligence, a high similarity score between a claim and a product feels like evidence and is not. In AI agent reputation, we measured a public registry and found eleven addresses writing two-thirds of the feedback in it. In county risk scoring, we published the weak three-month number beside the strong six-month one.
Here it is nine instead of twelve. Printing twelve would have been easier, would have matched what you have read elsewhere, and would have meant we had verified nothing. Nine is what our own evidence supports, and the three-state gap is in the post rather than buried in a footnote — because you would have found it the first time you looked one up.