A 25% figure, and not much underneath it
The gambling machine tax is going to 25%, according to reporting on the budget delivered by John Healey, with the additional take pointed at relief for the high street. That is the whole of the confirmed story at this stage: a headline rate, a stated purpose, and a date on the news wire of 28 October. SBC News carried the figure; the detail that decides who actually pays it sits in the budget documents, not the speech.
Everyone reading this already knows the instinctive response is correct. Machine taxes are rising, land-based venues will feel it, and the money will not stay where the press release says it goes. What most of the commentary gets wrong is why. This is not primarily a harm-reduction measure dressed up as fiscal policy. It is a revenue measure chosen because gambling machines are one of the few tax bases in Britain that cannot get on a plane.
Machine games duty, and why the band matters more than the rate
Machine games duty is a tax on the net takings of gaming machines on UK premises, paid by the person who holds the premises licence or the machine operator, not by the player at the point of play. It replaced the older amusement machine licence duty and VAT treatment in 2013, and it is charged in bands according to the maximum stake and prize a machine offers.
That band structure is the part that makes “25%” ambiguous rather than self-explanatory. HMRC’s published rates have run as three tiers:
| Band | Broadly covers | Published rate |
|---|---|---|
| Lower rate | Low-stake, low-prize machines | 5% |
| Standard rate | Most machines in arcades, pubs and betting shops | 20% |
| Higher rate | Machines with the largest maximum stakes | 25% |
A top band of 25% already exists. So a budget headline of 25% can mean one of two quite different things: the standard rate is being lifted from 20% to 25%, which would hit the bulk of machines in ordinary venues, or the higher band is being widened so that more machine types fall into it. The first is a five-point increase on the mass of the estate. The second is a reclassification that could bite harder on specific cabinets while leaving most pub and arcade machines where they are. Reporting of the speech does not settle which, and until the Finance Bill text and HMRC’s rate tables are published, nobody outside the Treasury can model it properly.
Two other things are not yet confirmed from the reporting: the commencement date, and whether remote gaming duty moves alongside it. Online casino play is taxed separately, under remote gaming duty on operator gross gaming yield, and a machine duty change does not automatically touch it. Treat any figure for the effective date or the revenue yield that you see quoted without a reference to the budget red book as someone’s guess.
The high street relief link is a framing device, not plumbing
The additional gambling duty revenue is not locked to high street relief in any legal sense. UK tax receipts go into the Consolidated Fund, and genuine hypothecation, where a specific tax is ring-fenced to a specific programme, is rare in British public finance and generally resisted by the Treasury precisely because it reduces flexibility. What a budget can do is announce a tax rise and a spending commitment in the same breath and let the arithmetic imply a connection.
That connection is doing real political work even if it is not doing accounting work. Retail relief, typically delivered through business rates discounts for shops, hospitality and leisure premises, is expensive and recurring, and it has to be paid for out of somewhere. Pairing it with a levy on gaming machines produces a tidy story: money from one kind of premises on the high street funds support for the rest of it. The story is cleaner than the mechanism. If machine receipts come in below forecast, the relief does not shrink; the gap is filled from general revenue like everything else.
Why gambling duty keeps getting picked
Governments reach for gambling levies because the base is immobile, the collection is already built, and the political cost of raising them is lower than almost any alternative. A gaming machine sits in a licensed premises with a known operator and audited net takings. There is no credible threat to relocate a fixed-odds terminal to Malta. Compare that with corporation tax or capital taxes, where behavioural response and mobility blunt the yield, and the appeal is obvious.
Then there is the second-order argument, which ministers rarely make explicitly but which does the heavy lifting: a tax on gambling can be defended as discouraging the activity even if it is being levied to fund something unrelated. Any duty increase can be presented either as fiscal housekeeping or as public health, depending on the audience, and the same measure can be both at once. That dual-use quality is what makes gambling taxation such a convenient instrument, and also what makes it easy to raise repeatedly without a coherent framework behind it. A rate set to fund retail relief has no particular reason to be the rate that best reduces harm.
The bill lands on fixed premises first
Machine games duty falls on land-based operators: adult gaming centres and seaside arcades, bingo clubs, betting shops, and pubs with machine entitlements. Those are businesses with rent, staff, energy costs and, often, thin per-machine margins. A five-point duty increase on net takings is not absorbed by changing a line in a spreadsheet; it is absorbed by removing machines, shortening hours, or closing marginal sites. Arcade and single-site operators have the least room to move, because they cannot cross-subsidise machine income from an online book.
Online operators are not directly touched by a machine duty change, which creates an awkwardness the trade bodies will press hard: a measure sold partly as support for the high street increases the cost of operating on it relative to operating from a server. Detailed industry reaction to this specific announcement was limited in the initial reporting, and it is worth waiting for the formal responses from the land-based trade associations rather than assuming their position from past campaigns.
What could go wrong with the numbers
The risk that matters for revenue forecasting is behavioural. Machine duty is charged on net takings, so if venues respond by reducing machine counts or closing, the tax base shrinks and the projected yield does not arrive. Operators can also adjust the product mix toward machines in cheaper bands, which is a rational response to a banded tax and a predictable erosion of the estimate. Costs can be passed through in the form of tighter returns or fewer promotions, though machine return-to-player ranges are themselves constrained by licence conditions, which limits how far that can go.
There is a harm argument on both sides and it should not be overstated in either direction. Higher duty can push operators toward higher-yield machine configurations, which is not obviously good for players. It can equally shrink the physical estate, which reduces exposure. The honest position is that the net effect on gambling harm from a duty change of this kind is not something anyone can state with confidence in advance, and framing a revenue measure as a protective one invites a conclusion the evidence does not support.
For players, none of this changes the underlying arithmetic: every gaming machine carries a built-in house edge, and losses over time are the expected outcome, not the unlucky one. UK venues are required to offer self-exclusion and staff are expected to intervene where play looks harmful. Anyone who recognises a problem in their own play should use those tools and the national support services rather than wait for a policy change to alter the odds, because it will not.
The documents that will settle this
Three things will convert the headline into something you can plan against. The Finance Bill text will show which band moves and whether any machine categories are reclassified. HMRC’s updated rate guidance will give the commencement date, and machine duty changes have historically applied from the start of an accounting period rather than on budget day, which matters for operators mid-period. The Office for Budget Responsibility’s costing will show what the Treasury actually expects the measure to raise, and that figure is the one to compare against receipts a year later.
Watch, too, for whether remote gaming duty is revisited separately. If it is not, the gap between the tax burden on a machine in Blackpool and a slot on a phone widens, and the next round of lobbying writes itself. The more interesting question is whether the tax-for-relief pairing gets reused. It polls well, it is administratively easy, and it worked once. That is usually enough.