Las Vegas Sands Q2 2026 revenue was flat at $3.15B but diluted EPS fell 19.7% to $0.53 as unusually low VIP hold and higher payroll and marketing costs in Macao cut property EBITDA 16% to $1.12B.
Revenue
$3.2B
-0.7% YoY
Net income
$346M
-24.9% YoY
Diluted EPS
$0.53
-19.7% YoY
Operating margin
19.6%
Q2 2026: Earnings fall as bad luck at the VIP tables and rising costs in Macao outweigh higher gambling volumes
Las Vegas Sands had roughly the same revenue in the second quarter of 2026 as a year earlier, $3.15 billion versus $3.18 billion (-0.7%), but earned much less from it. Operating income fell 21.1% to $618 million, net income attributable to the company fell 24.9% to $346 million, and diluted EPS fell 19.7% to $0.53 from $0.66.
Two separate things are going on, and it is worth keeping them apart:
Luck. A casino's revenue depends on how much customers bet (volume) and on what share of it the house keeps (the "win" or "hold" percentage). Over thousands of hands the win rate is predictable, but high-stakes VIP play can swing it sharply within a single quarter. In Macao this quarter, VIP ("Rolling Chip") win was far below normal. The Plaza Macao and Four Seasons actually posted a negative VIP win rate of -1.15%, meaning VIP customers there won money overall.
Costs. Even after adjusting for luck, profit still went down. In Macao, LVS is spending more on payroll, marketing and rewards to customers to hold its share of a competitive market. At Marina Bay Sands in Singapore, costs rose too, and a higher gaming-tax tier kicked in a month earlier than last year.
Volumes were healthy nearly everywhere. The 10-Q says table-game and slot volumes rose "across each of our properties" in Macao, and at Marina Bay Sands. The problem was turning that play into profit.
Key figures
Metric
Q2 2026
Q2 2025
YoY Change
Net revenues
$3,154M
$3,175M
-0.7%
Operating income
Read 0 community reports on Las Vegas Sands, or write your own.Write a report
Hold-adjusted property EBITDA (our calc. from company figures)
$1,169M
$1,247M
-6.3%
Macao adjusted property EBITDA
$430M
$566M
-24.0%
Marina Bay Sands adjusted property EBITDA
$689M
$768M
-10.3%
Macao mass-market table drop (5 properties)
$7,188M
$6,251M
+15.0%
Marina Bay Sands mass-market table drop
$2,597M
$2,360M
+10.0%
Marina Bay Sands RevPAR
$939
$844
+11.3%
Adjusted property EBITDA is the company's main measure of how profitable each resort is. It is earnings before interest, tax, depreciation, and corporate, development and pre-opening costs. Drop is the cash and chips customers bring to the table, so it measures how much mass-market (non-VIP) customers are playing. RevPAR is revenue per available hotel room. Operating margin (the share of revenue left after running the business, before interest and tax) is calculated from the income statement. The hold-adjusted EBITDA line adds management's estimated hold impact (from the Exhibit 99.1 earnings release) to reported EBITDA.
Takeaway: Bad VIP luck in Macao explains only part of the 16% EBITDA drop. On management's own hold-adjusted figures, property EBITDA still fell about 6% while mass-market drop grew 10-15%. LVS is buying volume in Macao with higher payroll and customer rewards, and that is squeezing margins in a way that luck will not reverse.
Separating luck from the underlying trend
The earnings release estimates what EBITDA would have been if VIP play had hit its expected win rate (3.3% in Macao; 4.2% this year and 4.1% last year at Marina Bay Sands):
Hold impact on adjusted property EBITDA
Q2 2026
Q2 2025
Macao (low hold this year = would add)
+$87M
-$7M
Marina Bay Sands (high hold both years = would subtract)
-$37M
-$80M
Total
+$50M
-$87M
This shows something the headline numbers hide. Marina Bay Sands was not unlucky this quarter. Its VIP win rate of 4.74% was still above the 4.2% it would normally expect. It just wasn't as far above normal as last year's 5.26%. The 10-Q attributes the Singapore EBITDA decline to "decreases in table games win percentages," which is accurate compared with last year, but compared with normal, Singapore still benefited from luck. Adjusted for hold on both sides, EBITDA was:
Macao: about $517M vs. $559M, -7.5%
Marina Bay Sands: about $652M vs. $688M, -5.2%
Consolidated: about $1,169M vs. $1,247M, -6.3%
So roughly two-thirds of the $215M reported EBITDA decline is luck, and about a third comes from the business itself.
Macao: more players, thinner margins
Macao net revenues were $1,776M, down just $11M from last year. Adjusted property EBITDA fell 24.0% to $430M. That puts the EBITDA margin (our calculation, EBITDA ÷ segment net revenue) at about 24.2%, down from 31.7%.
What drove it, per the 10-Q:
Volume was strong. Mass-market drop rose 15.0% across the five properties. VIP (Rolling Chip) volume rose about 73% to $7.57B, led by The Londoner Macao (+68.6% to $3.52B) and The Plaza/Four Seasons (+101.9% to $2.82B).
VIP win was very poor. The Venetian Macao won 0.62% of VIP volume (vs. 3.57%), and the Plaza/Four Seasons won -1.15% (vs. 2.72%). The Plaza/Four Seasons' casino revenue fell 51.6% even though its VIP volume doubled, and its EBITDA fell 69.7% to $20M. Mass-market win at The Venetian Macao also dropped to 20.4% from 23.5%.
Costs rose faster than revenue. Macao casino expenses rose $83M. The 10-Q attributes this to a $30M increase in gaming taxes, $26M in higher payroll "due to increases in table game hours and the competitive environment in Macao," and $11M in more casino marketing. Macao's credit-loss provision (money set aside for gambling debts that may not be repaid) rose $8M.
Renovation disruption. The Venetian Macao's room revenue fell 14.0% to $43M because rooms were taken out of service for the renovation that began in Q1 2026. Its EBITDA fell 30.1% to $165M, the largest decline of any property in dollar terms.
The Londoner Macao was the bright spot on revenue: casino revenue up 10.7% to $548M. Even so, its EBITDA fell 6.3% to $192M.
For the first half as a whole, management describes Macao this way: net revenue up 11.1% but EBITDA down 3.5%, "offset by increased costs on patron reinvestment and increased payroll costs related to the competitive environment." That pattern holds in both quarters, not just Q2. Revenue grows, profit doesn't.
The Macao subsidiary Sands China reported net income down 50.0% to $107M (vs. $214M).
Marina Bay Sands: still the profit engine, with less luck this time
Marina Bay Sands in Singapore produced $689M of adjusted property EBITDA on $1,378M of net revenue. That is a margin of about 50.0% (our calculation), down from 55.3%, and still roughly double Macao's.
Casino revenue fell 4.1% to $1,024M. VIP win fell to 4.74% from 5.26% and mass win to 22.9% from 23.7%, while mass drop rose 10.0% and VIP volume rose 3.6%.
Gaming tax tier hit a month early. Singapore taxes premium play at 8% until an annual threshold is reached, then at 12%. The 10-Q notes MBS "met the threshold in June 2026 versus July 2025," so June carried the higher rate this year but not last year. Along with payroll (+$9M) and higher gaming taxes (+$6M), MBS casino expenses rose $24M.
Non-gaming kept growing. Room revenue rose 12.7% to $151M. ADR (average daily room rate) was $982, up 10.6%, and occupancy was 95.6%. The 10-Q credits "the May 2025 completion of extensive renovations to introduce world class suites." Food and beverage rose $14M with a new venue opened in July 2025.
Costs below the casino line. General and administrative expense at MBS rose $24M: payroll +$7M, property taxes +$5M, repairs and maintenance +$5M, marketing +$4M. The credit-loss provision rose $12M.
Below the operating line
Development spending fell to $43M from $69M, mostly digital gaming ($35M vs. $49M) and Texas ($4M). Last year's figure included $19M for New York and Texas together. Management excludes this from adjusted EPS, which is one reason adjusted EPS fell more than GAAP EPS (-25.3% vs. -19.7%).
Interest income fell to $31M from $42M because there was less US cash after buybacks, and because the $1.26B seller-financing loan from the Las Vegas property sale was repaid early in May 2026. That cash came in, but the 4.25% interest it earned stops.
Tax rate rose to 19.1% from 14.8%. The earnings release attributes the rate mainly to the 17% Singapore rate, while Macao gaming profits are exempt from corporate income tax through 2027, so a quarter where Macao earns less pushes the blended rate up. The rate assumes Macao will renew the shareholder dividend tax agreement for 2026-27. LVS requested the extension in January 2026, and the 10-Q says "there is no assurance such agreement will be granted."
Buybacks softened the EPS decline. Diluted shares fell to 656M from 696M (-5.7%). That is why EPS fell 19.7% while net income fell 24.9%.
Capital returns and balance sheet
LVS repurchased $787M of stock in Q2 (about 15M shares at an average $52.37) and $1.53B in the first half. It paid $400M in dividends in the first half. In July the board raised the buyback authorization, which had fallen to $29M, to $6.0B through July 2029. It declared a $0.30 quarterly dividend and said it expects to keep that level through the rest of 2026. At June 30, unrestricted cash was $3.38B, debt (excluding finance leases) was $15.11B, and $4.26B of revolving credit was undrawn. LVS refinanced its $1.0B 3.500% notes due August 2026 with new senior notes issued in May.
Capital spending was $526M in the first half, including $317M at Marina Bay Sands. The large MBS Expansion Project is now expected to finish construction around June 2030 and open around January 2031. The Singapore government's deadline is July 8, 2029, so an extension would need its approval.
Outlook and our read
Management gave no numerical guidance. Its only forward statements are the dividend level, the enlarged buyback, and continued investment in "premium suites and other hospitality offerings" in Macao, including the ongoing Venetian Macao renovation. Q3 results are due in October.
Our read:
Macao's VIP luck should improve. A -1.15% VIP win rate at one property is statistical noise, and management's $87M hold estimate would come back if win returns to normal. But the underlying Macao problem is not about luck. On a hold-adjusted basis, EBITDA fell about 7.5% on 15% higher mass drop. As long as LVS keeps reinvesting in players and adding staff to defend share, Macao margins are likely to stay well below the low-30s% of a year ago.
Marina Bay Sands is the steadier asset but faces tougher comparisons. Its room and dining growth is real and comes from renovated product, but its VIP win ran above the expected rate in both this quarter and the year-ago quarter. If win drops back to the 4.2% theoretical rate, Singapore EBITDA growth would slow further. Higher G&A and property taxes point the same direction.
Per-share results depend more and more on buybacks. A $6.0B authorization is about $2B a year through mid-2029. With share count already down 5.7% year on year, per-share figures will look better than company-level profit for some time. Investors should judge the business by property EBITDA, not EPS.
Watch items for Q3: Macao hold-adjusted EBITDA margin (on our calculation, hold-adjusted EBITDA was about 27% of hold-adjusted revenue this quarter vs. about 31% a year ago; does it stabilize or keep falling?); disruption from the Venetian Macao room renovation; whether Macao renews the dividend-tax agreement; and MBS VIP hold versus the 4.2% theoretical rate.
Source: Las Vegas Sands Form 10-Q for the quarter ended June 30, 2026 (filed July 24, 2026), plus hold-adjusted and adjusted-EPS figures from the company's Q2 2026 earnings release (Form 8-K Exhibit 99.1, filed July 22, 2026). Property EBITDA margins and hold-adjusted EBITDA totals are our calculations from the company's reported figures.