Compact Revenue Sharing Compared: Arizona, Connecticut, Oklahoma

Three states offer a useful way to understand how tribal-state compact revenue sharing differs across the country: Arizona, Connecticut and Oklahoma. Each collects hundreds of millions of dollars a year from tribal casinos, yet each does it differently, and the differences reveal more about state bargaining power, market structure and political priorities than the headline percentages suggest. This analysis compares the three models side by side and explains what each implies for tribal governments and state budgets.

Federal law frames all of them. The Indian Gaming Regulatory Act does not allow a state to tax tribal gaming, so any payment must be negotiated in a compact and is generally supported only when the state grants something of value in return. The Legal Guide and the site's explainer on compact revenue sharing cover that doctrine. What follows is how three states have put it into practice.

Three formulas, three revenue bases

Arizona uses a graduated rate on Class III net win. Five large tribes pay 1 percent on the first $25 million, 3 percent on the next $50 million, 6 percent on the next $25 million and 8 percent above $100 million, while all other tribes pay a flat 0.75 percent. The proceeds are split by formula, with 12 percent going to local governments and 88 percent to the Arizona Benefits Fund. Our Arizona explainer walks through the fund's distribution.

Connecticut uses a flat share of a single product. The Mashantucket Pequot and Mohegan tribes contribute 25 percent of video slot receipts to the state, an arrangement dating to 1993 and 1996. The state expected about $365 million from slots this fiscal year, of which roughly $52.5 million is allocated to 169 municipalities. The contribution base excludes table games and non-gaming revenue. The accompanying piece on Connecticut's town aid dispute covers the politics.

Oklahoma uses an exclusivity fee under a model compact that any tribe may sign. Published accounts describe rates of roughly 4 to 6 percent on electronic games and up to 10 percent on table games, rising with revenue, with proceeds directed largely to education. Payments reached more than $221 million in the 2025 budget year according to the Oklahoma Indian Gaming Association. The Oklahoma model compact explainer details the renewal clause that currently runs the agreements through 2034.

What the differences mean

The first difference is the fee base. Arizona assesses Class III net win across all games, Connecticut only slots, and Oklahoma separates electronic games from table games. A rate cannot be judged without its base. A modest percentage applied to a large base can yield more than a high percentage on a narrow one, and a rate applied to net win captures casino margin rather than hotel or restaurant income.

The second is the rate structure. Arizona's scale ties the state's share to casino growth automatically, which tends to reward the state when large properties expand. Oklahoma's graduated tiers do something similar within each game category. Connecticut's flat 25 percent is simple and predictable, but it does not adjust for scale, and it makes the state's receipts depend on a single revenue stream. In a market where slot play is under pressure from changing player habits, as discussed in our look at the aging slot floor, that concentration carries risk.

The third is the destination of the money. Arizona earmarks nearly all of it by statute, with education and trauma care receiving about three-quarters of the Benefits Fund in the most recent quarterly breakdown. Oklahoma directs most of its proceeds to education through a dedicated revolving fund. Connecticut sends revenue into the state's general budget and then allocates a discretionary portion to towns, which is why the municipal share has swung from $88.1 million in 1993-94 to $52.5 million today. Earmarking provides a durable constituency; discretionary treatment invites annual budget conflict.

The most important variable is not the percentage but the exclusivity behind it. States can sustain high or growing payments only so long as the tribes continue to receive something scarce in return.

The exclusivity question underlies all three

All three models depend on tribal control of a defined market. That is why developments such as sports betting legislation, unlicensed online products and federal preemption arguments carry budget consequences for states as well as tribes. If a competitor captures wagers that would otherwise be placed in the tribal channel, the fee base shrinks. The Michigan experience, in which tribal payments to the state fell sharply as gambling expansion tested exclusivity, is examined in the report on the 71 percent drop in Michigan payments, and it illustrates the same dynamic from the opposite direction.

For tribes, the lesson is that a revenue-sharing agreement is only as strong as the exclusivity that anchors it, and that renewal and renegotiation clauses deserve as much attention as rates. A tribe that agrees to a payment tied to a protected market has leverage when that market is threatened, and several compacts now include provisions for reduced or suspended payments if exclusivity is breached.

For states, the lesson is that high rates are not free. They are justified by the value of what the state grants, and an aggressive push for higher percentages without a corresponding grant risks litigation or stalled negotiations. The 2015 GAO review found that, of 276 compacts examined, 107 involved no revenue sharing, 100 paid between 10 and 14.9 percent, and 22 paid between 15 and 25 percent, a distribution that shows how widely arrangements vary.

Readers can use the comparison tools to explore how other states' tribal gaming markets differ, and the state hubs for Arizona, Connecticut and Oklahoma for operator-level detail.