Three closed venues in a quarter when revenue went up
Rank Group shut three venues last quarter, in Swansea, Sunderland and at Grosvenor Casino Reading Central, during a period when group revenue grew five per cent to £216.1m. That combination is the whole story in miniature, and it is why Machine Games Duty explained properly matters more than most tax explainers: a business can be growing at group level and still have individual sites where the arithmetic has stopped working.
Chief Executive Richard Harris was blunt about what he thinks happens next. Any increase to Machine Games Duty, he said, “will increase costs, reduce venue viability, lead to venue closures and job losses, reduce tax receipts and halt investment.” He added that HM Treasury had recognised those economic arguments the previous year, and that with the right conditions Rank’s casinos and bingo clubs could support the Government’s growth agenda.
That is an operator making a lobbying case ahead of a Budget, and it should be read as one. But the mechanism he is describing is real, and it is worth understanding on its own terms.
What is Machine Games Duty?
Machine Games Duty (MGD) is a UK tax on the net takings from gaming machines in premises such as casinos, bingo clubs, adult gaming centres, pubs and arcades. It is charged on what the machines keep, meaning money staked minus prizes paid out, rather than on turnover or on a venue’s overall profit.
It is a narrow lever, not a general gambling tax. Betting shops and online operators pay different duties on different things, which matters when you hear “gambling tax rise” in a headline and have to work out who is actually exposed. A change to MGD lands on the machine line of a land-based venue’s P&L and nowhere else.
What counts as a machine game under the duty
A dutiable machine game is one played on a machine where the prize can be money, or money’s worth, and the outcome depends to some extent on chance. Slots and the fixed-odds electronic terminals you find on a casino floor sit squarely inside that definition. Games that only pay out non-cash prizes of limited value, skill-with-prize machines and certain lottery and bingo products are treated separately, which is one of the reasons operators care a great deal about how a machine is classified.
How is Machine Games Duty calculated?
MGD is charged as a percentage of net takings in an accounting period, with machines sorted into bands by how much a single play costs and how big the prize can be. HMRC operates a lower rate of 5%, a standard rate of 20% and a higher rate of 25%, applied to the net takings from the machines in each band rather than to the venue’s total income.
In broad terms, the lower band covers low-stake, low-prize machines where a play costs no more than 20p and the maximum prize is no more than £10. The higher band covers machines where a single play costs more than £5. Everything in between falls into the standard band. Operators register with HMRC, file returns for their accounting periods and pay on the net figure, so a quarter with unusually generous jackpot payouts genuinely does produce a smaller duty bill.
Both the rates and the stake and prize thresholds that sort machines into bands are set in legislation and can be changed at a Budget. Check HMRC’s Machine Games Duty guidance for the figures in force before applying them to anything.
Why Rank pointed at the Budget rather than at trading
Rank’s argument is that the threat of a higher MGD rate, not just current trading, is shaping decisions about which venues stay open. Harris said speculation about the upcoming Budget “continues to weigh heavily on investor sentiment,” and framed the three closures as the start of a pattern rather than a one-off clean-up.
Here is the part worth sitting with. Rank’s land-based estate grew five per cent in the quarter, and online gaming grew six per cent. This is not a company describing collapse. It is a company saying that marginal sites have no cushion, and that a tax change would move several of them from just about viable to not worth keeping open. Those are different claims, and only the first is evidenced by the numbers in the trading update. The second is a forecast made by a party with an obvious interest in the outcome.
The honest position for a reader is that both things can be true at once. An operator can be overstating the cliff edge for negotiating purposes, and the cliff edge can still exist for individual venues.
Can a tax increase really force a casino to close?
Yes, when the venue’s operating margin is thinner than the duty increase applied to its machine income. A duty rise does not reduce a casino’s rent, staffing rota or licensing costs, so the entire increase comes out of whatever profit the site was making, and at some sites that profit is small enough to be wiped out.
Fixed costs do not flex
A land-based casino carries a heavy fixed cost base: property, business rates, utilities, security, compliance staff, dealers, licence fees. Those costs barely move with footfall. A venue that takes 10% less in a quarter does not get to pay 10% less rent. That is the structural difference between a physical casino and an online operation, and it is why the same group can report online growth while shutting buildings.
Why machines, not tables, drive the margin
Machines are the high-margin line in most land-based venues. They need no dealer, run continuously, occupy little floor space per unit of revenue, and the hold percentage is predictable. Table games are labour-intensive and spiky. So when a duty is levied specifically on machine net takings, it is hitting the part of the business that pays for the part that does not pay for itself.
The following arithmetic is illustrative and not reported figures. Say a venue produces £4m of standard-rate machine net takings a year. At 20%, MGD costs £800,000. Raise the rate to 25% and the bill becomes £1m, a £200,000 increase. If that venue’s annual operating profit before the change was £150,000, it is now loss-making, and no amount of cost trimming on a site that is already running lean closes a gap of that size. Change the assumed profit to £900,000 and the same rise is an irritation rather than an existential problem. The rate matters less than the margin it lands on, which is exactly why closures cluster at specific sites instead of spreading evenly across an estate.
How MGD differs from the UK’s other gambling duties
MGD is one of several distinct gambling duties, each with its own base and rate. Knowing which one a policy story is about tells you immediately which operators are exposed.
| Duty | What it taxes | Typically paid by | Published rate |
|---|---|---|---|
| Machine Games Duty | Net takings from gaming machines on premises | Casinos, bingo clubs, AGCs, arcades, pubs | 5%, 20% or 25% by machine band |
| Remote Gaming Duty | Operator’s gaming profits from remote casino play | Online casino operators | 40% of gaming profits (increased from 21% on 1 April 2026) |
| General Betting Duty | Bookmaker’s profits on general bets | Bookmakers, retail and remote | 15% of net stake receipts for general bets |
Spread betting and pool betting sit under their own rates within the betting duties, and casino table games are covered by a separate banded gaming duty rather than by MGD. The practical point is that an increase to Remote Gaming Duty and an increase to Machine Games Duty hit entirely different buildings, balance sheets and payrolls. Treating them as one “gambling tax” is how readers end up misjudging which part of the industry a Budget actually touches.
What happens if Machine Games Duty rises further?
A further MGD rise would concentrate its effect on the weakest land-based sites rather than spread thinly across the sector. Rank has already said more closures would follow, and the same logic applies to independent adult gaming centres and smaller bingo operators that lack a group balance sheet to absorb a hit.
The plausible responses are limited and none of them is costless. Operators can close marginal venues, which removes jobs and tax receipts along with the duty. They can reconfigure machine mixes toward lower-rate bands, which usually means a weaker product and less revenue per machine. They can consolidate, buying estates cheaply and stripping overhead. They can lobby, which is what Rank is doing in public. Or they can shift investment toward online, where the cost base is lighter, which is the outcome least likely to serve a high street.
Rank’s counter-argument is that a rise would “reduce tax receipts”, and that is the genuinely unsettled question. Whether a higher rate on a shrinking base raises more or less than a lower rate on a stable one depends on how many venues actually close and how much of their spend migrates elsewhere rather than disappearing. Nobody outside the Treasury’s modelling can answer that from the published figures, and the Treasury’s answer and the operator’s answer have never matched.
Reading the next gambling tax headline properly
Three questions get you most of the way through any gambling tax story. What is the rate, before and after. What base is it applied to: turnover, net takings, gross gaming yield or profits, because the base decides how brutal a given percentage is. And who absorbs it, meaning whether the payer has the margin to swallow an increase or is already operating close to break-even.
Run Rank’s statement through that filter and it becomes legible. The lever is MGD, the base is machine net takings, and the party absorbing it is a venue estate with high fixed costs and uneven site-level profitability. That does not prove the closures Harris predicts are inevitable, and his incentive to predict them is plain. It does mean the mechanism he describes is not rhetorical. Whether a government should protect marginal venues through the machine duty, or accept their closure as the price of a higher rate, is a political judgement rather than an arithmetical one, and the figures in a quarterly trading update will never settle it.
