For more than a decade, all stakeholders involved in online gaming were facing a high degree of legal uncertainty in Austria. Demand was substantial and growing, yet the law reserved online casino and lottery products to a single concessionaire operating under the win2day brand. Everyone else served Austrian players from abroad, in a grey zone that generated a steady stream of player restitution litigation but left the monopoly with a shrinking market share of about 30-40%. The draft amendment now in consultation sets out to end that situation and to replace the monopoly with an open, licensed online market. It is the most significant recalibration of Austrian gaming law in a generation.

From monopoly to licensing

The core move is straightforward. Instead of a single online concession, the draft creates a licensing regime under which multiple operators can be authorised to offer online gaming in and into Austria, subject to strict player protection, supervisory and tax conditions. A dedicated independent gaming supervisory authority is to be established to run the system. Until that authority is operational, the transitional provisions keep the competent tax office, Finanzamt Österreich, in charge of issuing concessions. The existing monopoly concession is set to expire in September 2027, which is the pivot around which the whole transition 
is organised.

The reform keeps faith with the objectives that have always underpinned Austrian gaming law and that the Court of Justice of the European Union has repeatedly held must justify any restriction on the freedom to provide services: player protection, the prevention of gaming-related crime and the containment of the black market. The draft adds an explicitly stated ambition that is worth quoting in substance, because it becomes the benchmark for everything else. The regime is meant to achieve a high level of channelisation into the regulated market by offering an attractive legal product, while maintaining the highest possible standard of player protection.

That single sentence carries a lot of weight. Channelisation is not one objective among several. It is the precondition for all of the others. A deposit limit, a (self-)exclusion register or a monitoring obligation protects only those players who actually play in the licensed market. Every player who stays with, or drifts to, an unlicensed operator is beyond the reach of all of it. Any measure in the draft, therefore, has to be assessed twice: once for its player-protection value, and once for its effect on channelisation. Where the two pull in opposite directions, the reform has a design problem to solve.

The player protection framework

The draft assembles a comprehensive set of measures for player protection. Mandatory deposit limits are one economic part: a maximum of €250 ($285.80) per week for players up to the age of 26, and €1,680 per month for players from 26 onwards. Crucially, the law also allows a higher, individualised limit for players from the age of 23, where there is no indication of harm and graduated additional safeguards such as monitoring and feedback tools are in place. This is a sensible, proportionate structure that recognises that not every economically capable adult is a vulnerable player.

For online gaming offered in the form of slot machines, the draft imports the elements familiar from the land-based world: a maximum stake of five euro per game, a maximum prize of €10,000 per game, a minimum game duration of two seconds and a mandatory cooling-off. These limits do not apply to other game formats, which is good news, as this takes into account the fact that, for example, poker is a completely different game in terms of betting/winnings and game structure and cannot be treated with the same limits.

It also introduces two cross-operator registers. A blocking register consolidates self-exclusions and third-party exclusions across all licensees and gaming forms. A limit register records and enforces deposit limits across operators, closing the obvious loophole of a player simply opening accounts elsewhere.

Alongside these sits a Safe-Server requirement, under which relevant gaming data is held in a form that allows the supervisory authority transparent, auditable and integrity-assured access. Taken together, overall it is a modern, credible player-protection architecture that stands comparison with any regime in the European Union.

Also the fiscal side of the reform deserves a note, because it changes commercial planning materially. Under the new bonus rules, promotional credits, free spins and similar incentives count as stakes for tax purposes. Licensees are permitted to deduct bonuses only within capped limits. Refunds, including payments made on the basis of civil-law invalidity of a gaming contract, do not reduce the tax base. Operators modelling the Austrian market will need to price these mechanics in from the outset, as they directly affect the economics of player acquisition.

The questions practice still has to answer

None of the above is the difficult part. The difficult part is a small number of design choices where the draft, as it stands, risks working against the channelisation objective it sets for itself. Four of these stand out from an advisory perspective.

The timing of the cooling-off. The transitional regime effectively requires prospective applicants to cease their offering by a fixed date, with a lengthy blocking period as the penalty for continuing. Yet the date from which a license can actually be obtained is not fixed anywhere. Neither the tender, nor the procedure’s duration, nor the moment of the first license grant is settled. The expiry of the incumbent’s concession in September 2027 is not a guarantee that new licenses will be live by then. Operators are therefore asked to withdraw from the market on 1 January 2027 in exchange for an uncertain return, leaving the entire online market temporarily to the former monopolist and a possible black market.

The Netherlands offers a cautionary precedent: a cooling-off period imposed before market opening pushed a significant share of demand towards operators who never intended to seek a license, and that demand had to be won back afterwards. A cleaner design would attach the cooling-off to conduct rather than to a calendar date, so that it functions as a sanction for those who decline to enter the regulated framework, not as a precondition for entry that punishes the very operators willing to be licensed. Cooling-off, in other words, should follow from a lack of regulatory willingness, not stand in front of it.

Stake and prize limits: the right scope, and the case for keeping the figures adjustable. The draft deserves credit for how it scopes these limits. The five euro stake and €10,000 prize caps apply only to online gambling in the form of slot machines, while other games such as poker remain subject to the general player-protection rules. That is a sound, risk-based call: a poker tournament, with a single buy-in and prizes drawn from a participant pool, does not carry the risk profile of a high-frequency slot game. The level of the slot limits derives from the already liberalized and regulated land-based slot industry and a strict line on high-frequency products is legitimate.

The narrower question is where such figures belong. Limits of this kind benefit from regular review against channelisation and harm data, which suggests they may sit more comfortably in an implementing ordinance than in primary legislation, without any loss of stringency. Germany, where stake limits are fixed in the State Treaty, illustrates how difficult adjustment becomes at that level. Implementation matters too: international studios certify their products for many jurisdictions at once, and parameters that depart substantially from prevailing technical norms require Austria-specific versions that some suppliers may be reluctant to implement. Neither point argues for less strict player protection, but only for parameters that can be calibrated, monitored, revisited and guarantee high channelisation at the same time.

The deposit-limit increase in practice. The statutory possibility of raising the limit for suitable players is welcome, but its value will be decided by its implementation. If the increase process is quick, fully digital and workable, it channels high-value players into the supervised environment. If it is slow, paper-bound or effectively unreachable, it does the opposite, driving exactly those players to unlicensed sites where no limit, no monitoring and no protection exist at all. The ordinance should mandate clear, standardised and automatable processes in the limit register.

Calibrating to the market, not to the incumbent. A recurring temptation in reforms of this kind is to take the incumbent’s commercial parameters and impose them on every new licensee. That would be a mistake here. The reform rests on the finding that the monopoly product did not channel demand adequately despite its exclusivity. Parameters that failed to produce an attractive offer under monopoly conditions will not succeed under competition. The benchmark should be a genuine market standard, an offer that a competent, non-vulnerable player experiences as a real alternative to the black market.

A reform worth getting right

Austria is doing the right thing – and doing most of it well. The structures are modern, the player-protection instruments are serious, and the direction of travel is plainly correct. The reservations set out above are not objections to the reform but to a handful of transitional and technical choices that could blunt it. Whether the new market channels demand from its first day will turn less on the ambition of the framework, which is not in doubt, than on the practicality of these final design decisions. That is where the consultation phase, and the ordinances still to come, will earn their keep. 

This article was written while the public consultation on the draft amendment was still open, with a submission deadline of 15 July 2026. Individual provisions discussed here may therefore have been amended in the course of the legislative process by the time of publication.