The US$1.72 billion figure is a ceiling, not a bill
A study summarised by the Asian gaming trade outlet GGRAsia projects that South Korea’s casinos could lose up to US$1.72 billion in gaming spend once MGM Osaka opens. That number is the top of a modelled range for spend diverted away from Korea, not a forecast of confirmed losses, and treating it as the latter is the most common misreading in the current conversation about casino market competition in Asia.
The distinction matters commercially. “Up to US$1.72 billion in diverted gaming spend” describes money that Japanese and other regional players might wager closer to home instead of in Incheon, Seoul or Jeju. It is not the same as US$1.72 billion wiped off Korean operators’ gross gaming revenue, because spend at the table is what players put through, while GGR is what the house keeps after payouts. Anyone building that figure into a model without reading the original study’s base year, source-market split and hold assumptions is importing someone else’s assumptions as if they were facts.
What the study claims, and what it leaves open
The headline claim is a range with US$1.72 billion at the upper end, tied to the opening of MGM’s Osaka integrated resort. The projection rests on a condition almost nobody repeats alongside the number: that today’s travel patterns, visa rules and player preferences broadly persist, and that a meaningful share of Korea-bound Japanese gamblers were travelling because Japan had no legal casino, rather than because they wanted Korea specifically.
The Osaka resort is currently targeted to open around 2030 on the Yumeshima waterfront, which means the projection is being argued over several years before it can be tested. Large IR timelines slip. Construction costs, regulatory conditions and partner structures all move, and a five-year-out revenue diversion estimate carries wider error bars than its two decimal places suggest.
What the available summary does not settle: how the study splits exposure between the VIP gaming segment, the premium mass market and ordinary leisure visitors. That split decides everything downstream, because those three groups behave nothing alike when a new option appears.
Why Osaka hits Korea harder than it hits anyone else
Korea’s foreigner-only casinos are structurally dependent on inbound demand that Japan is about to serve domestically. Korean nationals cannot legally play at them, which is the point: the business was built on Japanese, Chinese and other visiting players. Remove the Japanese segment’s main reason to board a flight and you remove a chunk of demand that cannot be replaced with local walk-ins.
Three mechanics compound:
- Geography. Osaka is a domestic trip for tens of millions of Japanese residents. A short-haul international flight plus passport control loses to a train ride.
- Scale. MGM is a global operator with an existing premium database across Macau and Las Vegas, and the Osaka project is a full integrated resort rather than a hotel casino floor.
- Segment overlap. The travellers most sensitive to convenience, the higher-frequency premium mass players, are exactly the ones Korea’s operators have been courting since junket-driven VIP volumes thinned out across the region.
The honest counterpoint is that Japan’s IR regime includes entry restrictions and fees for domestic residents. That friction is real and it blunts the effect. It does not come close to the friction of international travel.
What an integrated resort actually sells
An integrated resort combines a casino floor with hotels, MICE facilities, retail, dining and entertainment venues, so that gaming revenue is only one line in the P&L. Singapore’s two IRs and Macau’s Cotai properties both run substantial non-gaming businesses, and Japan’s licensing framework was written around conference and tourism outcomes rather than gambling for its own sake.
That mix is why a single new IR can redraw regional flows rather than simply adding capacity. A conference delegate, a family on a theme-park weekend and a concert attendee all arrive for non-gaming reasons and some of them gamble while they are there. A standalone casino has to win the trip on gaming alone. When a mega-resort with 3,000 rooms and an exhibition hall opens 500 kilometres from an older, smaller property, the older property is not losing a gambling argument. It is losing a trip-planning argument.
Scale also buys the things cross-border gamblers notice: more table variety, deeper limits, suites and jets for the top end, and a loyalty currency that is worth something on the next trip. Smaller operators can match the hold percentage. They cannot match the itinerary.
Which hubs are genuinely exposed, and which are not
Exposure depends on how much of a hub’s revenue comes from travellers who have a nearer substitute. On that test, the regional picture is uneven.
| Hub | Core demand base | Main competitive pressure |
|---|---|---|
| South Korea (foreigner-only) | Visiting Japanese and Chinese players; no domestic play permitted | Highest. Its primary source market gains a domestic option |
| Kangwon province (locals-permitted) | Korean residents | Low. Domestic demand is unaffected by an overseas opening |
| Macau | Mainland Chinese visitors, premium mass and VIP | Moderate. Policy, concession investment duties and mainland travel rules dominate over Osaka |
| Singapore | Regional leisure and MICE, premium mass | Moderate. Competes for the same Asian premium traveller, with reinvestment underway at both IRs |
| Entertainment City, Philippines | Domestic mass plus regional and Korean visitors | Mixed. Mass growth has been offsetting a thinner VIP segment |
Macau is often lumped into the same sentence as Korea here, and it should not be. Its swing factors are mainland policy and visitor flows, not a Japanese resort. The Philippines story is more about internal mix than foreign competition, which is visible in how Okada Manila’s mass gaming growth offset a 21.4% VIP decline. Korea is the outlier because its exposure is specifically to the loss of a source market, not a shift in player type.
The response options, ranked honestly
Four levers get discussed. They are not equally credible.
- Non-gaming investment. The strongest structural answer and the slowest. Building reasons to visit Korea that have nothing to do with the casino floor is the only lever that changes the trip-planning argument, and it requires capital years before the threat lands.
- Source-market diversification. Realistic and already happening. Southeast Asian and other regional visitors reduce Japanese concentration risk, though they typically spend differently per head.
- Loyalty and VIP reinvestment. Useful for defending the known high-value players an operator can actually identify, which is why casino loyalty technology that follows the player off the floor keeps attracting investment. It protects a database; it does not create one.
- Regulatory easing. Periodically floated in Korea, including questions about domestic access. Treat any version of this as discussion, not decision, until legislation exists.
My verdict: marketing spend and VIP comps will absorb most of the near-term budget because they show results inside a quarter, and they will be the least effective of the four. Convenience cannot be bought back with points.
What would confirm or kill the projection
Watch three things before 2030. First, whether Osaka’s construction and licensing milestones hold to the current timetable, because every year of slippage is another year of Korean revenue the study implicitly writes off. Second, the Japanese share of visitation and drop at Korea’s foreigner-only properties, which is the direct denominator in the study’s logic. Third, the final shape of Japan’s domestic entry rules and fees, since heavier friction for local residents would push some of that demand back out to regional hubs.
If Japanese visitation to Korean casinos is already flattening before Osaka pours concrete, the study was conservative. If Korean operators have converted their floors toward Chinese and Southeast Asian premium mass by then, the US$1.72 billion headline will read as a scenario nobody got around to disproving. The number is a hypothesis with a date attached, and 2030 is when it gets marked to market.