Policy risk is not an operator problem you can ignore
A tax rise on bookmakers does not stop at the bookmaker. That is the myth worth killing first, because it shapes how most people read stories like the one running in the UK right now. Gambling tax policy risk is the chance that a government changes the rate, the base or the timing of gambling taxation faster than operators can adjust, and the adjustment always travels: into margins, then into staffing and shop estates, then into the odds, the promotions and the games a player can actually access.
Two live examples make the pattern visible. In Britain, the Betting and Gaming Council has gone into the Budget arguing that more than 36,000 betting shop jobs are exposed to a further duty increase. In India, the GST Council’s decision to tax real-money gaming at 28 per cent on full face value rather than on operator revenue did something more drastic in a shorter time. Different countries, different instruments, same mechanism.
What the UK is actually arguing about
The specific fight is over Machine Games Duty, not gambling tax in general. According to the BGC’s pre-Budget intervention, independent modelling by EY suggests that raising MGD to 40 per cent could cost up to 16,000 jobs, put almost 1,500 betting shops and as many as 34 casinos at risk, and leave the Treasury up to £124 million worse off than before the rise.
That last number is the whole argument in miniature: the industry is not claiming the tax is unfair so much as claiming it is self-defeating, because a shrinking taxable base cancels out the higher rate. Whether that holds depends entirely on assumptions about how many shops close and how much of their turnover simply disappears rather than moving elsewhere.
The workforce figures the BGC puts alongside it are worth reading as advocacy with real data behind it. More than 80 per cent of betting shop jobs sit outside London, in market towns, former industrial areas and coastal towns. Nineteen per cent of betting shop employees are under 25 against 11 per cent across the wider economy, and 51 per cent are under 35 against 35 per cent nationally. Women make up half the workforce. All of these are the trade body’s own figures, presented to influence a fiscal event, and a careful reader should treat the 36,000 headline as an industry claim to verify rather than an audited fact. That does not make it wrong. It makes it an estimate produced by an interested party.
Why a betting shop cannot absorb a tax rise the way a supermarket can
Retail betting runs on high turnover and thin net margin, so a few points of duty consumes a large share of the profit on each shop. A licensed shop carries fixed costs that do not flex with a tax change: lease, staff rotas, machine licensing, compliance and safer-gambling obligations, security, cash handling. There is very little discretionary spend left to cut once those are paid.
The result is that estate decisions are made shop by shop at the margin. A unit clearing a small profit becomes a loss-maker the moment duty moves, and it closes. That is why gambling tax debates produce job-loss numbers in a way that, say, a one-point corporation tax change does not.
Turnover or gross gaming revenue: the design choice that decides the damage
How a gambling tax is calculated matters more than the headline rate. A tax levied on gross gaming revenue takes a share of what the operator keeps after paying out winnings. A tax levied on turnover or on the full amount a player deposits or stakes takes a share of money that mostly gets recycled back to players as payouts.
The difference is not academic. On a market with a 5 per cent operator margin, a 10 per cent GGR tax costs half a point of that margin. A 10 per cent turnover tax costs twice the entire margin. The same headline percentage can be a cost of doing business or an extinction event depending only on the base it is applied to. This is the single most useful thing an informed player can understand about betting industry taxation, and it explains why operators fight hardest over the base and only secondarily over the rate.
India’s 28% GST on real-money gaming: the clearest precedent
India moved the base, not just the rate. The GST Council decided in 2023 that online real-money gaming would attract 28 per cent GST calculated on the full face value of deposits or entry amounts, replacing an earlier treatment under which many skill-gaming platforms paid 18 per cent on platform fees or gross gaming revenue. Readers should confirm current specifics against the GST Council’s own notifications and CBIC circulars, since the position has moved more than once and remains the subject of litigation.
Two features made the Indian change more severe than anything the UK is currently debating. First, the shift from commission to face value multiplied the effective burden several times over for formats where the same money is staked repeatedly. Second, tax authorities pursued the new interpretation for past periods, issuing retrospective demands that were reported in the Indian press as running to figures far larger than the sector’s annual revenue. A prospective rate rise is a business problem. A retrospective reinterpretation of the base is a solvency problem, and it is the reason the Indian case escalated into the Supreme Court rather than staying a lobbying exercise.
What operators and players actually felt
Operators responded by raising the cut taken from each contest, cutting marketing, laying off staff and in some cases shutting down formats entirely. Several smaller platforms exited; larger ones consolidated. Players saw it as worse prize pools, smaller bonuses and fewer low-stake tables, because when the tax lands on deposits there is no way to hide it in the back end.
Government revenue from the sector did rise after the change, which complicates the industry’s “self-defeating tax” argument. What nobody has credibly measured is how much Indian play simply moved to offshore sites that collect no GST at all. That gap matters, and I will come back to it.
Two fights, one mechanism
The structural comparison is short enough to lay out directly.
| UK Machine Games Duty debate | India RMG GST change | |
|---|---|---|
| What is taxed | Gaming machine profits in licensed premises | Full face value of deposits or entry amounts |
| Level under discussion | Up to 40% in the modelled scenario | 28% |
| Retrospective element | No | Yes, via demands for past periods |
| Headline industry claim | Up to 16,000 jobs, ~1,500 shops, 34 casinos at risk; Treasury £124m worse off (EY modelling cited by the BGC) | Effective tax burden multiplied; platform closures and layoffs reported |
| Status | Pre-Budget lobbying | Implemented, litigated, followed by further restrictive legislation on online money games |
Read across the rows and the verdict is fairly plain. The UK argument is about a rate on an already-taxed base, which is survivable even if painful. India’s was about redefining the base and then applying the new definition backwards, which is not. If you have to choose which kind of change to fear, fear the base change and the retrospective one. A phased, pre-announced rate rise lets operators reprice, renegotiate leases and plan closures over quarters. A sudden reinterpretation forces all of that into weeks, and compliance cost rises on top of the tax itself.
How a tax line reaches your odds and your bonus
The chain is short and it is not speculative. Duty rises, net margin per customer falls, and operators restore margin with the levers they control: trimming promotional budgets first, because bonuses are the most discretionary spend in the business; then pricing, meaning slightly worse odds, higher rake or a bigger overround on the same market; then product, meaning low-margin formats and low-stake tables get withdrawn; then estate and headcount.
Players rarely see a notice explaining any of this. They see a welcome offer get smaller, a familiar market priced a little tighter, a shop on the high street shuttered. The gambling regulation impact arrives as a worse deal, not as an announcement.
The further consequence is migration. When the licensed price of play deteriorates, some demand moves to offshore and grey-market sites that pay no local duty, run no local safer-gambling obligations and offer no recourse when a withdrawal is refused. That is the outcome regulators say they least want, and the one a tax rise can produce without anyone intending it.
Where the evidence runs out
Nobody has a reliable number for how many players move offshore after a tax rise. This is the hole in the middle of every argument on both sides, and it is worth being blunt about it rather than filling it with a plausible-sounding percentage.
Industry-commissioned models have to assume a migration rate to generate their job-loss and revenue-loss figures, and the assumption drives the result. Treasury forecasts have to assume one too, usually a smaller one. Offshore activity is by definition unlicensed and unreported, so neither side can measure the thing their case depends on. The EY figure cited by the BGC and any official costing of the same measure can both be internally consistent and still differ by an order of magnitude because of this single unknown.
So the honest position is that there is a revenue-maximising rate somewhere, that pushing past it damages the tax base and the regulated channel together, and that nobody can tell you with confidence where it sits in either country. Treat anyone who claims precision here with suspicion, whoever is paying them.
Signals worth tracking before the next change
Policy risk is forecastable in direction even when it is not forecastable in size. The markers below give a few months’ warning in most markets.
- Budget and fiscal-event calendars, plus any pre-Budget consultation on gambling duty structure, which is where a base change would first appear in writing.
- GST Council meeting agendas and CBIC notifications for Indian RMG taxation, including anything touching valuation rules rather than rates.
- Trade body commissioning of economic impact modelling, which almost always precedes a measure the sector expects to lose on.
- Operator disclosures about shop closures, estate reviews or market exits, which usually land in results statements before they land on the high street.
- Litigation progress, since a court ruling on the tax base can reverse or entrench a change more decisively than any budget.
- Licensing and payment-blocking enforcement activity, a rough proxy for how much play has already moved offshore.
Watching those lets you read a headline about 36,000 jobs for what it is: one side’s estimate in an argument where the central number is unknowable, and an early signal that the terms of play in a regulated market are about to shift. If you gamble, the practical response is unglamorous. Know the licensing status of the operator holding your money before a policy shock makes the offshore alternative look cheaper, and keep your own deposit and loss limits set regardless of what the tax rate happens to be.
