Cboe Global Markets and the Expensive Tollbooth
What the business is
An exchange is a toll bridge that sometimes benefits from storms. Cboe Global Markets earns transaction fees when investors trade, while market data and connectivity subscriptions provide a steadier stream when trading activity subsides.
Cboe is the largest US options-exchange operator. Its crown jewels are S&P 500 index options and VIX derivatives. The SPX franchise has an exclusive licence through 2032 and a non-exclusive arrangement through 2033, while Cboe owns the VIX methodology. More than 98% of US index-option volume is executed on Cboe-affiliated exchanges.
That intellectual property is reinforced by a network effect. Liquidity attracts orders, deeper order books support tighter spreads, and tighter spreads attract more liquidity. A competitor can build a matching engine; persuading institutions to abandon an established market is the less cooperative part. Regulatory approval and clearing infrastructure create further barriers.
Options generated about 61% of net revenue in 2025. Data and Access Solutions contributed about 26%, largely through recurring subscriptions. The rest includes equities trading in North America, Europe and Asia, futures and global foreign exchange. Transaction revenue responds to volume, especially during volatile markets, while data and access revenue provides some ballast.
The moat appears to be widening through short-dated products such as 0DTE options and VIX1D, longer Global Trading Hours, approval for EDGX to offer 23x5 equities trading, and continued development of recurring data revenue. Algorithmic and AI trading do not obviously threaten the exchange model. More automated trading may increase demand for execution and data without weakening the licensed products.
Competition is sharper in multiply listed options, where IEX, MEMX and MIAX can pressure fees and margins. That is worth watching, but it does not reproduce the SPX and VIX complex. The more consequential risk is the licensing horizon: unfavourable terms after 2032 or 2033 could weaken the legal foundation of the moat itself.
What the numbers say
Annual owner earnings are approximately $1,151.4 million. The thesis treats owner earnings as relatively close to net income because exchange operations require modest capital expenditure. I regard that as a reasonable approximation, not a complete distributable-cash analysis.
Cboe is not chiefly a lending business, so loan growth is not the central capital constraint. Even so, regulated exchanges and clearing infrastructure require resilience, and the available evidence does not reconcile regulatory-capital needs, operating safeguards or other restrictions on cash available for dividends and repurchases. That uncertainty argues against treating every dollar of estimated owner earnings as immediately distributable.
The screening snapshot reported a five-year average return on equity of 16.0% and debt-to-equity of 0.28. Those figures indicate respectable returns without conspicuous reliance on leverage.
Capital allocation has been mixed. The Bats Global Markets acquisition brought technology integration and European reach. The later ErisX purchase produced a $460 million impairment and an effective retreat from Cboe Digital. Cboe has paid dividends for 16 consecutive years and had roughly $680 million remaining under its repurchase authorization, but the acquisition record still warrants a raised eyebrow when management considers another large transaction.
The present valuation arithmetic is less complicated:
- Market cap (price-adjusted, 2026-10-09 close): $30.90 billion
- Owner-earnings yield (price-adjusted): 3.73%
- Share price (2026-10-09 close): $304.16
- 10-year Treasury: 5.24%
The resulting spread is negative by 1.51 percentage points. In other words, the equity offers a 3.73% owner-earnings yield against a 5.24% Treasury before allowing for operational, regulatory and licensing risks.
The thesis used a reference method of capitalizing owner earnings at 12 times under low growth and 15 times under structural growth. The stronger case rests on continued 0DTE adoption, broader participation, international expansion and recurring data revenue. Those are understandable assumptions, but the valuation evidence provides no usable intrinsic-value range for the rubric, so circle is unavailable. I keep that framework separate from the directly observable yield comparison rather than granting precision to a reference value I cannot independently verify here.
A fine tollbooth can remain a fine tollbooth while its shares become expensive. Markets are capable of holding both thoughts at once.
What Wall Street believes vs what I see
There is no live consensus estimate in the evidence. The available headlines nevertheless outline the prevailing debate: one presents Cboe as a beneficiary of round-the-clock trading, another warns that record results depend on elevated activity, and a third reconsiders the case after a CFTC decision. The mood is appreciative of the franchise but conscious of volume and regulatory risk.
I share the appreciation. Licensed products, concentrated liquidity and recurring data revenue form a durable combination. Global Trading Hours and short-dated options can deepen that franchise, while automated trading may reinforce demand for both execution and information.
Where I differ is mostly in the price paid for those virtues. A 3.73% owner-earnings yield against a 5.24% Treasury assumes that growth and durability will compensate for the lower current yield. That may happen, but the arithmetic offers little shelter if trading volume normalizes, regulation constrains 0DTE products, licensing terms deteriorate or capital allocation relapses.
I would monitor two consecutive quarters of double-digit declines in SPX or VIX average daily volume, changes involving the S&P DJI licence, regulatory proposals affecting 0DTE products, and any acquisition above $1 billion. The market sees an unusually good tollbooth. I see the same tollbooth and a rather ambitious toll charged for owning it.
Verdict
Verdict: pass.
The 3.73% owner-earnings yield is below the 5.24% Treasury, failing the rubric’s watch threshold. Circle is unavailable because there is no usable intrinsic-value range with which to apply the required discount test. A yield above the Treasury, supported by durable earning power rather than elevated trading activity, would change the verdict.
This analysis is AI-generated, educational, and not investment advice. Figures may contain errors or be delayed. Disclaimer