Intercontinental Exchange from a Long-Term Business Owner's Perspective

ICE is not a single exchange, but a market infrastructure group made up of three major segments. In 2025, ICE's net revenue came from Exchanges ($5.411 billion), Fixed Income and Data Services ($2.419 billion), and Mortgage Technology ($2.101 billion). Exchanges includes futures, options, equities, and clearing networks, with NYSE as the best-known asset; Fixed Income and Data Services provides data, indices, valuations, execution, CDS clearing, and related services; Mortgage Technology provides origination, closing, servicing, and data analytics software to lenders, servicers, and housing-finance participants.

Its customer base is not mysterious either: customers in the exchange segment include dealers, brokers, market makers, institutional investors, listed companies, and clearing members; the fixed-income and data business serves asset managers, banks, trading-platform users, and data subscribers; the mortgage technology business serves loan originators, servicers, settlement participants, and software users across the housing-finance chain. The company's monetization mainly falls into three categories: trading and clearing fees, subscription/data/index fees, and workflow software fees. Data and software revenue is naturally more recurring, while trading-related revenue is more driven by market volatility and transaction volume.

In terms of "revenue quality," ICE's biggest advantage is not an unusually hot trading quarter, but a revenue structure that has formed a combination of "recurring revenue + upside sensitivity to volatility." In Q1 2026, the company disclosed in investor materials that overall recurring revenue grew 6% year over year, while transaction revenue grew 30% year over year; annual subscription value, or ASV, in Fixed Income and Data reached $2.036 billion; and of Mortgage Technology's Q1 2026 revenue of $539 million, recurring revenue was $401 million. This means that even if market trading activity cools, ICE still has a sizable revenue base that looks like "infrastructure rent."

On cost structure, ICE has typical platform-company characteristics: it builds the front-end platform and runs the back-end network, with low marginal cost but meaningful fixed cost, technology development, depreciation, and amortization. In 2025 operating expenses, compensation and benefits were $1.963 billion, technology and communications were $870 million, and depreciation and amortization were $1.560 billion, showing that this is a business of "heavy software, heavy networks, and heavy intangible-asset amortization," not a business of heavy inventory or heavy manufacturing. For that reason, gross margin is of limited reference value, while operating margin and cash-flow conversion matter more.

Key dependencies should be assessed in layers. ICE does not clearly depend on a single customer, nor does it carry inventory risk or single raw-material supplier risk; but it depends heavily on regulatory permissions, market-structure position, customer workflow embedding, core technology stability, and key executives. Jeffrey Sprecher remains one of the company's most important strategic figures. This "founder-like influence" is both an advantage and a source of key-person risk.

If the stock market were closed for 5 years, I would still be broadly willing to own ICE, provided I accepted that it is not a consumer-staples company to hold without thinking. It is steadier than a traditional cyclical financial company, more protected by regulation than a pure software company, but also more complex than a Coca-Cola-style consumer business. It is a good business that is understandable, but not ultra-simple.

Business understandability score: 4.5/5. For long-term investors, ICE's core is not complicated: it owns a set of hard-to-replace financial infrastructure nodes and continuously monetizes the data, workflow, and clearing capabilities of those nodes.

III. Industry and Competitive Landscape

ICE's industry is more accurately described as the financial market infrastructure industry. This is not a high-explosion emerging sector, but a high-barrier track within a mature industry: exchanges, clearing, indices, data, pricing, fixed-income workflow, and mortgage software. Long-term demand across the industry is stable because global capital markets, risk management, collateral management, price discovery, and financing activity will not disappear; what really changes is transaction volume, the rate environment, data-distribution methods, and regulatory requirements.

Long-term industry demand is stable, but different subsegments have different cyclical properties. Exchanges and clearing often make more money during high-volatility periods; data and workflow are more like "subscription rent"; mortgage technology is more affected by interest rates and the real-estate cycle. In October 2025, the Mortgage Bankers Association expected U.S. single-family mortgage originations in 2026 to grow by about 8% to $2.2 trillion, with both purchase and refinancing activity rebounding; but Reuters reported in May 2026 that MBA weekly data still showed U.S. mortgage rates around 6.56%, and high rates suppress applications. This is exactly why ICE's mortgage tech has recovery upside and macro vulnerability at the same time.

On major competitors, the strongest competitor in exchanges and clearing is CME Group; on equities, options, and data, competitors include Nasdaq and Cboe; and in global data and post-trade infrastructure, LSEG must also be considered. Based on current market pricing, ICE's trailing P/E of about 22.3x is below CME's 24.8x, Nasdaq's 27.4x, and Cboe's 30.5x. This shows that the market recognizes ICE as a high-quality asset, but has not given it the highest valuation premium among peers.

ICE's industry position can be summarized as: a strong global derivatives and energy/rate risk-management platform, the owner of NYSE, an important player in fixed-income data and workflow, and one of the core U.S. mortgage technology suppliers. In March 2026, the company disclosed that its monthly futures and options volume set a record, with monthly contract volume of 428.9 million, more than 70% above the previous record; Q1 2026 presentation materials also showed 23% year-over-year growth in futures & options open interest. This indicates that ICE's trading ecosystem is still strengthening, not deteriorating.

Is the industry easy to disrupt technologically? The answer is that "parts will change, but the foundation will not change easily." Matching, clearing, compliance records, index licensing, and data distribution are deeply embedded in regulation and customer processes, and cannot be overturned by a new app; however, single-point products in data pricing, workflow software, electronic bond trading, and mortgage tech will still face competitive pressure from more agile startups, large software companies, and customers' in-house systems. Regulatory change is equally important. For example, the SEC's adjustment to the implementation dates for U.S. Treasury central clearing rules shows that market infrastructure businesses remain under continuous policy reshaping.

Overall, ICE is not an "excellent company in a bad industry," but closer to a "good company in a good industry": high industry barriers, concentrated profit pools, persistent long-term demand, losers being eliminated, and winners growing stronger.

Industry attractiveness score: 4/5. The deductions come mainly from regulation, technology evolution, and the cyclicality of mortgage tech, not from structural weakness in the industry itself.

Moat and Management

IV. Moat Analysis

ICE's moat is not "brand mindshare" in the consumer sense, but an institutionalized moat.

Start with the strongest elements. First is regulatory licenses and infrastructure status. Exchanges, clearinghouses, market data, and certain key benchmark businesses are naturally protected by licenses, rules, customer access, and regulatory oversight, making replication extremely difficult. Second is network effects and a liquidity moat. Once liquidity forms in futures, options, equities, and CDS clearing, customers tend to keep trading in the deepest, most stable, and cheapest pools, and liquidity itself attracts more liquidity. Third is switching cost. Once fixed-income data, indices, pricing, and mortgage servicing software enter customer systems, compliance processes, and reporting chains, replacement is not only troublesome but can also create regulatory and operational risk. Fourth is a combined data and workflow advantage. ICE does not merely sell matching; it also sells data, analytics, indices, workflow, and settlement, allowing it to earn multiple layers of revenue from the same customer.

Scale also matters. The company had 2025 net revenue of $9.931 billion and Q1 2026 TTM adjusted EBITDA of $7.042 billion, showing that it is a giant platform capable of continuously investing in technology, compliance, acquisitions, and buybacks. Scale brings not only cost advantages, but also the ecosystem advantage that "everyone wants to connect to the main node." Replicating the full ICE is not a matter of "spending several hundred million dollars to build a system"; it would require years or even more than a decade of heavy capital, regulatory permissions, core liquidity, and embedding in customer workflows.

Does it have pricing power? Broadly yes, but pricing power is not exercised through crude price hikes. It is monetized through index licensing, data pricing, software seats, workflow expansion, functional upgrades, and higher penetration across trading bundles. In Q1 2026, Fixed Income & Data ASV was $2.036 billion, clearly above $1.883 billion a year earlier; Mortgage Technology's Origination Tech revenue grew 10% year over year. This shows that even amid inflation and a complex environment, ICE has not lost monetization ability.

Can it remain profitable in an economic downturn? Historically, probably yes. Part of its revenue is positively correlated with market volatility, while another part is data and software subscriptions; this is the "all-weather business model" management repeatedly emphasizes. In 2025, the company's GAAP operating margin was 50%, and operating cash flow was $4.662 billion; in Q1 2026, when market volatility increased, it had even stronger upside sensitivity. The point to watch is that mortgage tech is the real drag, not the core exchange/clearing/data network.

My judgment on ICE's moat is: overall stable and moderately widening. The core moat in exchanges and clearing is stable; data and fixed-income workflow are widening; the mortgage tech moat is relatively less deep and more affected by the rate cycle, centralized customer purchasing, and competition.

Moat strength score: 4.5/5. It is not 5 because ICE is not completely immune to cycles, and not every business line is equally solid; mortgage tech and acquisition integration in particular still need ongoing validation.

V. Management and Capital Allocation

Jeffrey Sprecher's significance to ICE is closer to that of an "institutional founder" than a hired professional manager. He has served as chairman and CEO for a long time, giving the company strong strategic continuity. The 2026 proxy shows that 9 of the company's 11 director nominees are considered independent directors, and the company has an active Lead Independent Director mechanism; it also clearly mentions stock ownership retention requirements for executives and directors. This indicates that the governance structure is not a low-standard model in which "the CEO alone decides everything."

But in terms of ownership alignment, management is not an extreme owner-operator with a very large personal stake. According to the 2026 proxy, Jeffrey Sprecher beneficially owned about 3.481 million shares, while all directors, director nominees, and executives together owned about 4.782 million shares, both below 1% of total shares outstanding. In other words, management and shareholders are not badly misaligned, but this is also not a case where "executives have their entire net worth tied to the company like Berkshire managers." You can give credit for governance and culture, but cannot automatically assign a perfect score just because the company has founder color.

On capital allocation, ICE's long-term model is clear: use strategic M&A to build the platform first, then use strong cash flow for deleveraging, dividends, and buybacks. In 2025, the company returned $2.4 billion to shareholders, including $1.3 billion used for buybacks; in Q1 2026 it repurchased another $551 million of shares. In February 2026, the company also raised its quarterly dividend by 8%, implying an expected annual dividend of $2.08 per share. This shows that management has not abandoned capital discipline.

The question is: Do acquisitions continue to create value? This is the most important point in evaluating ICE's management. Acquisitions such as Ellie Mae and Black Knight did expand ICE from an exchange group into a broader financial technology platform; but they also pushed the company toward substantial goodwill and accumulated intangible assets. By Q1 2026, ICE's goodwill had reached $30.634 billion, and other intangible assets, net, were $15.108 billion; together they were far above shareholders' equity of $29.478 billion. This means that past acquisitions by management were possibly right in commercial logic, but have also left a very high level of "intangible-asset risk" on the accounting and balance-sheet side. If mortgage tech falls short of expectations for a long time, the damage will first show up in asset impairment and valuation compression.

On buybacks, my view is "rational, but not especially brilliant." The company is indeed reducing share count: diluted weighted average shares declined from 576 million in 2024 to 575 million in 2025, and to 570 million in Q1 2026. But these buybacks look more like steady-state capital returns than aggressive countercyclical action during extreme undervaluation. In other words, buybacks are positive, but they should not be exaggerated into superbly timed capital allocation.

Overall, I think management is trustworthy, long-term oriented, and execution-capable, while capital allocation is closer to "excellent but imperfect" because external acquisitions have deepened the platform but also materially raised the balance sheet's intangible-asset share and valuation dependence.

Management and capital allocation score: 4/5. The positives come from strategic consistency, capital discipline, and governance; the deductions come from ownership that is not especially heavy, large post-acquisition intangible assets on the balance sheet, and buybacks that are not clearly countercyclical.

Financial Quality and Owner Earnings

VI. Financial Quality Analysis

Conclusion first: ICE's financial statements are generally of high quality, and profits broadly correspond to real cash profits. Earnings growth does not require heavy tangible capital investment; however, its balance sheet has one weakness that must be faced directly: goodwill and intangible assets account for a very high share. What this type of company fears most is not inventory write-downs, but the failure of acquisition logic.

The table below prioritizes key financial metrics that I directly confirmed from official materials. One note: to follow the discipline of "only write data with sources," I mark a few annual revenue figures that were not directly captured from original tables in this round as "not directly confirmed." This is not a flaw, but a deliberate choice to avoid fabrication. FCF in the table is operating cash flow minus capital expenditures and capitalized software development costs. 2020-2021 data mainly comes from the 2021 10-K, 2022-2024 data mainly comes from the 2024 10-K, 2025 data comes from the 2025 full-year earnings release, and TTM is stitched from full-year 2025 plus Q1 2026/Q1 2025.

This table shows several important things. First, 2022-2023 profits were clearly affected by acquisitions and integration, so earnings quality cannot be judged from a single year. Second, 2024-2025 cash flow recovered and rose to a very strong level. Third, ICE's growth is not a case of "the more it grows, the more cash it needs"; it still maintains very strong cash conversion. In 2025, FCF/net income attributable to ICE was about 1.17x, and TTM FCF/net income attributable to ICE was about 1.08x, showing that GAAP profit is not paper profit.

On margins, ICE's 2025 GAAP operating margin was about 49.6%, and net margin attributable to ICE was about 33.4%; 2024 operating margin was about 46.4%. This is not a cyclical stock's accidental high point, but closer to the structural advantage of a platform business with high fixed costs and low marginal costs. In 2025 operating expenses, depreciation and amortization reached $1.560 billion, indicating that historical acquisitions left a large amount of accounting amortization; this suppresses GAAP profit, but does not necessarily suppress cash flow at the same pace. The issue is that long-term investors cannot treat all amortization as "irrelevant" because if the company relies on ongoing acquisitions to maintain competitiveness, part of that amortization represents real economic cost.

Capital returns need to be viewed separately. Using a simple calculation, 2025 ROE was around 11%-12%; the company itself disclosed adjusted ROIC of 9% and adjusted cash ROIC of 11% for Q1 2026. It is important to emphasize that ICE's ROA is not very useful because clearing and margin activities significantly expand the asset base and dilute headline ROA. For this type of company, ROIC, cash return, and free cash flow per share are more meaningful.

On liabilities, headline liabilities are large, but a substantial part relates to customer margin and default funds in the clearing business and should not be confused with corporate debt. What really matters is corporate debt / EBITDA / interest coverage. As of Q1 2026, ICE reported debt of $20.370 billion, adjusted EBITDA of $7.042 billion, and debt/adjusted EBITDA of 2.9x; if only unrestricted cash of $863 million is netted, net debt/adjusted EBITDA was about 2.8x. In 2025, GAAP operating income was $4.929 billion and interest expense was $803 million, giving interest coverage of about 6.1x. This means the balance sheet is not fragile, but it is also not a "zero-leverage fortress."

On working capital, a service company has no inventory risk. In Q1 2026, accounts receivable rose from $1.552 billion to $2.382 billion, while customer margin and related receivables also fluctuated sharply. This was mainly related to active market conditions, clearing activity, and trading during high volatility, and should not be simply interpreted like manufacturing as "worsening collections." I care more that, in this volatile background, the company still generated $1.326 billion of quarterly operating cash flow.

The most important point to watch is weak tangible net-worth protection. As of Q1 2026, goodwill of $30.634 billion plus intangibles of $15.108 billion totaled about $45.742 billion, significantly above shareholders' equity of $29.478 billion. In other words, ICE has negative tangible net assets. This does not mean it is dangerous. A high-quality platform can absolutely be "light in tangibles, heavy in intangibles." But it does mean: if the competitiveness created by acquisitions is disproven, book assets will not give you much downside protection.

Overall judgment: ICE's profits are mainly real cash profits; growth does not require excessive tangible capital investment; I have not seen obvious signs of financial fraud or aggressive revenue recognition. But this company is a typical combination of "very strong cash flow and very heavy intangible assets." For long-term shareholders, the most important worry is not whether it will suddenly run out of cash, but whether a major acquisition later turns into an impairment.

VII. Owner Earnings Analysis

Using a Buffett-like Owner Earnings framework, I would stay conservative. Net income attributable to ICE in 2025 was $3.315 billion; Q1 2026 net income attributable to ICE was $1.413 billion, versus $797 million in the prior-year period. On a TTM stitched basis, net income attributable to ICE was about $3.931 billion. For non-cash expenses, TTM depreciation and amortization was about $1.555 billion; TTM stock-based compensation, roughly calculated as full-year 2024 plus Q1 2026 minus Q1 2025, was about $252 million.

Maintenance capital expenditure is the hardest part to confirm precisely because the company does not separately disclose maintenance capex. I take a conservative treatment here: treat all capital expenditures and capitalized software development costs as part of a "maintenance + growth" mix and deduct all of them first. Under this approach, TTM operating cash flow was about $5.022 billion, TTM capital expenditures plus capitalized software totaled about $778 million, and TTM reported free cash flow was about $4.244 billion. After further considering the economic dilution from stock-based compensation, I deduct about $252 million of SBC from FCF, arriving at conservative Owner Earnings of about $3.99 billion.

This matches intuition: ICE's real distributable cash flow is usually close to or even slightly above net income over the long term. In 2025, FCF/net income attributable to ICE was about 117%; under my more conservative Owner Earnings definition, TTM Owner Earnings/net income attributable to ICE was still around about 102%. In other words, this is not a fake high-quality growth company that appears to earn a lot but actually has all cash consumed by capital expenditures.

At a current market capitalization of about $87.19 billion, ICE implies:

• P/FCF of about 20.5x;

• P/conservative Owner Earnings of about 21.8x;

• Conservative Owner Earnings yield of about 4.6%.

These numbers are not bad, but for an investor who is "balanced and somewhat conservative, requiring a margin of safety," they are not attractive enough. The valuation looks more like a fair price for "a high-quality compounding machine" than a price cheap enough to blunt most errors.

Valuation and Margin of Safety

VIII. Intrinsic Value Estimate

Here I use three methods: Owner Earnings discounting, relative valuation, and asset/liquidation value. Core conclusion first: ICE is not obviously overvalued, but it is not obviously undervalued for conservative investors either.

Method One: Owner Earnings Discounting

My baseline is the conservative TTM Owner Earnings of $3.99 billion discussed above, with share count understood around the 566.4 million shares outstanding disclosed in the March 2026 proxy. Assumptions are as follows:

Converted to per-share values, this roughly corresponds to:

• Conservative intrinsic value: $125-135/share;

• Fair intrinsic value: $155-175/share;

• Bull intrinsic value: $195-210/share.

The most important point is not precision after the decimal point, but the conclusion: the current $152.97 is roughly above the conservative value and near the lower edge of fair value. This is the classic profile of "high quality, fair price, but not a thick margin of safety."

Distinguishing facts, assumptions, and inferences: Facts are the current market capitalization, cash flow, debt, and operating performance; assumptions are the next 10 years' Owner Earnings growth rate, discount rate, and terminal growth rate; the inference is "whether the company can keep widening its moat and maintain roughly 6% compounding." For ICE, the most fragile part of the valuation is not "how much it earns next year," but whether the market is willing to give it a cash-flow multiple above 20x for a long time.

Method Two: Relative Valuation

Based on current public-market pricing, ICE's trailing P/E is about 22.3x, below CME's 24.8x, Nasdaq's 27.4x, and Cboe's 30.5x. Using the latest company and peer balance-sheet data in a rough calculation, ICE's current EV is about $106.7 billion, corresponding to about 15.2x TTM adjusted EBITDA; for Nasdaq, using LTM EBITDA of $3.189 billion, total debt of $8.957 billion, and cash of $515 million disclosed in its Q1 2026 presentation materials, EV/LTM EBITDA is about 18-19x. Cboe's Q1 2026 adjusted EBITDA was $544.6 million, but because its first-quarter results were significantly boosted by volatility, directly annualizing EV/EBITDA would be too optimistic, so it is only a reference. In other words, ICE is not expensive relative to peers and is even slightly cheaper, but it is also not cheap enough to indicate an obvious mispricing.

This shows an important point: ICE's valuation must be understood in the context that high-quality exchange/market-infrastructure assets are generally not cheap. You cannot say ICE is cheap just because all peers are expensive; but you also should not ignore that it is indeed cheaper than some peers. For conservative investors, relative cheapness versus peers can reduce the risk of overpaying, but it does not automatically create a margin of safety.

Method Three: Asset or Liquidation Value

This is also where I am most cautious about ICE. As of Q1 2026, ICE had shareholders' equity of $29.478 billion; but goodwill was $30.634 billion, and other intangible assets were $15.108 billion. This means ICE's book value lacks hard-asset backing, and tangible net worth is negative. In other words, if you want to use "liquidation value" to find a floor, ICE is not a good candidate. Its value lies in networks, licenses, liquidity, customer embedding, and software ecosystems, not land, factories, inventory, or large net cash. The conclusion of the asset approach here is not "cheap," but "not suitable for proving cheapness with liquidation thinking."

Valuation Conclusion Summary

• Conservative intrinsic value range: $125-135/share

• Fair intrinsic value range: $155-175/share

• Bull intrinsic value range: $195-210/share

• Current price relative to fair value: near fair value, slightly neutral

• Current price relative to conservative value: above conservative value, no obvious margin of safety

• Ideal buy price range: $120-135/share

• Acceptable holding price range: $135-165/share

• Clearly overvalued price range: above $190/share, requiring more caution

IX. Margin of Safety

My judgment is clear: the current margin of safety is insufficient. This is not because ICE is a poor business, but because the market already recognizes that it is a good one. Using conservative Owner Earnings, the current price gives you an initial yield below 5%. If future growth slows a little, rates remain elevated, or the market's valuation multiple for high-quality financial infrastructure falls, you may still buy a good business that keeps growing, yet receive only unremarkable shareholder returns.

The most fragile valuation assumptions are mainly twofold. First, mortgage tech can recover enough to support the group's medium-term growth. Second, the market's quality premium for ICE will not materially compress. If growth over the next decade is only 3%-4%, not around 6%, then returns from buying today may be only slightly above 10-year Treasuries. On May 21, 2026, the U.S. 10-year Treasury yield was about 4.57%; ICE's conservative Owner Earnings yield was about 4.6%. The initial risk premium between the two is not rich.

So this is a classic case of: a good company at a price that is not bad enough to reject immediately, but not good enough for conservative investors to buy heavily with confidence. If you emphasize odds over certainty, it is worth waiting for a better price; if you emphasize business quality and long-term portfolio stability, a small position and long-term tracking are not unreasonable.

Risks, Counterarguments, and Investment Checklist

X. Risks and Counterarguments

The most important risk is not day-to-day volatility, but permanent capital loss. I group the most important risks into the following categories.

The first is competition and industry erosion risk. The core barriers in exchanges and clearing are strong, but fixed-income data, bond execution, and mortgage tech are not invincible. CME, Nasdaq, Cboe, LSEG, and more vertical software and workflow players are all competing for a larger share of wallet. If customers gradually split workflows, build internally, or shift to cheaper alternatives, ICE's recurring-revenue quality will be reassessed.

The second is interest-rate and mortgage tech risk. Although the Mortgage Bankers Association forecast U.S. single-family mortgage originations to recover to $2.2 trillion in 2026, Reuters reported that May 2026 MBA weekly data showed mortgage rates rising to 6.56% and weighing on applications. In other words, this business has expected rebound potential but depends heavily on the rate path, leaving limited room for forecast error.

The third is regulatory and market-structure risk. The profits and moat of the market infrastructure industry are largely built on rules, licenses, and market structure; rules can change. The SEC's delay of implementation dates for U.S. Treasury clearing rules shows that regulation is not static. Any change in market data pricing, clearing access, auction structure, capital requirements, or trading-fee disclosure could affect ICE's monetization model.

The fourth is M&A and goodwill risk. As of 2026Q1, ICE's goodwill and intangibles together exceeded $45.7 billion, higher than shareholders' equity. If integration goes well, this is not a problem; but if an important acquired franchise fails to meet expectations for a long time, impairment and valuation compression could both occur.

The fifth is overvaluation risk. Even if the company continues to execute well, if the long-term center of risk-free rates rises, the market may no longer be willing to assign cash-flow multiples above 20x to high-quality exchange-like assets. For new buyers today, this risk of "valuation not collapsing, just gradually compressing" is the easiest to overlook.

The strongest bear case can be summarized as: "ICE is indeed a good company, but the market has known the best parts for a long time. Exchange and data businesses are already mature, the recovery pace of mortgage tech is uncertain, and the balance sheet is full of intangible assets created by acquisitions. If you buy now, the result may not be bad, but it could be a decade in which 'the business is fine, while shareholder returns are ordinary.'" This view is not superficial; it is actually powerful.

Which investors might be bearish? Deep-value investors may question: why pay more than 20x Owner Earnings for a mature company with a large amount of intangible assets? Macro-oriented investors may say: high rates will suppress mortgage tech and valuation multiples. Industry-competition investors will focus on: Rocket's acquisition of Mr. Cooper, industry-chain restructuring, changes in customer budgets, and data/software competition. Interestingly, the shareholder-proposal opposition discussion in ICE's 2026 proxy also listed bear angles such as Rocket's acquisition of Mr. Cooper, management insider selling, slowing growth, and market-sentiment concerns.

What facts would overturn the investment judgment? If the following happen, I would significantly lower my long-term view of ICE: first, fixed-income/data and core exchange segments lose pricing power or market share for multiple consecutive quarters; second, mortgage tech still fails to regain growth and profitability during a falling-rate cycle; third, debt/EBITDA remains above 3.5x and cash returns do not improve; fourth, a major goodwill impairment or failed acquisition integration occurs; fifth, major regulatory changes weaken the structural advantages of its clearing, data, or trading ecosystem; sixth, management continues large-scale buybacks while per-share intrinsic value growth clearly slows.

The largest permanent capital-loss scenario is not a short-term decline in trading volume, but: mortgage tech value being overestimated, goodwill created by acquisitions being impaired, valuation multiples for the core trading/data businesses being revised downward, and high interest rates making the market unwilling to pay high multiples for a "high-quality but relatively mature" platform. Such an outcome may not break the company's operations, but it would keep your purchase cost from earning an ideal return for a long time.

XI. Comparison with Other Opportunities

Compared with its strongest competitor, CME, ICE's advantage is that its business is more diversified, spanning exchanges/clearing, data, indices, fixed-income workflow, and mortgage software. Its disadvantage is higher business complexity and greater volatility in mortgage tech quality. CME's current trailing P/E is about 24.8x. The market's payment for CME is closer to a "pure exchange leader valuation"; ICE's 22.3x is slightly lower, which to some extent reflects a more diversified business that is also harder to value simply.

Compared with Nasdaq, ICE trades at a lower valuation, while Nasdaq has a stronger growth narrative around financial technology transformation. Nasdaq's Q1 2026 presentation disclosed LTM EBITDA of $3.189 billion and gross leverage of 2.8x, indicating that this type of asset is also viewed by the market as a "high-quality compounding platform." ICE is not obviously inferior; it is simply more tilted toward "exchange + mortgage technology," with less pure-software valuation imagination.

Compared with Cboe, ICE's current P/E is also lower, while Cboe's business is more concentrated in options, volatility, and global equities markets. Cboe had Q1 2026 cash of $2.134 billion and debt of $1.443 billion, making its balance sheet look lighter; but its quarterly profit is also more sensitive to the volatility environment. ICE has greater breadth, but weaker hard-asset backing.

The comparison with the risk-free yield makes the issue sharper: the U.S. 10-year Treasury yield was about 4.57% on May 21, 2026, while ICE's current conservative Owner Earnings yield was also about 4.6%. This means buying ICE is really a bet on future cash-flow growth and valuation maintenance. If you are not confident in growth, ICE does not offer a particularly thick initial risk premium versus Treasuries.

Compared with broad-market indices, my conclusion is: ICE deserves a place on the watchlist, but it is not clearly superior to "just buying the index" enough to justify a large position today. If your portfolio could hold only 5 assets, does ICE deserve a slot? My answer is: in terms of business quality, it qualifies; in terms of current odds, it may not belong there right now.

XII. Investment Checklist

XIII. Final Investment Conclusion

【Final Rating】 Watch

【One-Sentence Investment Thesis】 ICE is a financial market infrastructure company with a strong moat, strong cash flow, and long-term compounding potential, but buying at the current price looks more like buying "certainty" than buying "cheapness".

【Core Bullish Reasons】 First, exchanges, clearing, data, indices, and workflow software form a compound platform with a deep moat. Second, operating cash flow, free cash flow, and Owner Earnings in 2025 and TTM prove high earnings quality. Third, the revenue structure combines a recurring base with upside sensitivity to high volatility. Fourth, long-term governance and capital allocation are generally rational, with dividends continuing to rise and buybacks continuing. Fifth, while the current valuation is not low, it is not aggressive relative to high-quality peers.

【Core Bearish Reasons】 First, the margin of safety is not obvious, and the conservative cash yield is already close to the 10-year Treasury. Second, mortgage tech is materially affected by interest rates and the real-estate cycle. Third, goodwill and intangible assets are extremely high, making book-value defense weak. Fourth, buybacks look more like steady-state returns, not highly reflexive buying at a deep discount. Fifth, if valuation multiples revert, you may get a "good business, ordinary return."

【Key Assumptions】 For the investment to work, at least the following must hold: the exchange and data businesses continue to maintain pricing power; mortgage tech at least does not drag down medium-term group growth; debt/EBITDA remains within a controllable range; there is no major acquisition impairment; and the market does not structurally revalue high-quality financial infrastructure assets downward.

【Fair Buy Price】 $120-135/share better fits the requirements of a "balanced and somewhat conservative" value investor. This range corresponds to around the conservative value in this report to a slight discount, leaving more cushion for assumption error.

【Target Holding Period】 More than 10 years. ICE's advantage lies in long-term compounding, not short-term catalysts.

【Expected Annualized Return】 At the current price, a rough estimate is:

• Conservative scenario: 4%-6%/year

• Base scenario: 7%-9%/year

• Bull scenario: 10%-12%/year

The main source of these returns is not dividends, but Owner Earnings growth and valuation maintenance.

【Maximum Downside Risk】 If mortgage tech recovery fails over the next few years, core-segment growth slows, acquisition-related intangible assets are impaired, and the market compresses the valuation multiple from the current level into a lower range, a 30%-45% medium- to long-term share-price decline is not unimaginable; if severe adverse regulation or major strategic mistakes occur, the decline could be even larger. The core issue is not bankruptcy risk, but permanent return damage caused by not buying cheaply enough.

【Tracking Indicators】 The 8 indicators most worth tracking going forward are: first, Exchanges net revenue and open interest/ADV; second, ASV growth in Fixed Income & Data; third, the split between recurring and transaction revenue in Mortgage Technology; fourth, operating cash flow and FCF; fifth, conservative Owner Earnings and per-share growth; sixth, debt/adjusted EBITDA; seventh, whether goodwill and intangible assets show impairment signals; eighth, whether per-share intrinsic value is truly growing after buybacks.

【Signals That Trigger Reassessment】 The investment logic must be reviewed if any of the following occur: competitive position in core derivatives and clearing worsens; data/ASV growth falls to low single digits with no signs of recovery; mortgage tech still fails to recover in an improving rate environment; debt/EBITDA deteriorates; a major goodwill impairment appears; regulation changes the economics of trading, data, or clearing.

【Final Recommendation】 Calmly put, ICE deserves respect and deserves a long-term place on a high-priority watchlist. It answers the question "is this a business I would be willing to own for a long time" with an answer leaning yes; but it has not sufficiently answered the question "can I buy today with an adequate margin of safety," where the answer is closer to not yet. If you already own it, I would lean toward continuing to hold and track it; if you are preparing to buy new shares, I would put discipline ahead of price and avoid treating affection for a good business as a direct reason to buy.

Open Questions and Limitations This report prioritizes company annual reports, quarterly reports, investor materials, and authoritative market data. Therefore, for a few 2022-2023 revenue figures not directly captured from original tables in this round, as well as some peer PB / P/FCF metrics, I deliberately did not add unconfirmed data. In addition, because the company does not separately disclose maintenance capital expenditure, it can only be handled with conservative assumptions, so Owner Earnings is a "conservative approximation," not an audited definition.

This report is based on public information and does not constitute investment advice. Markets carry risk; invest with caution.