This is a business I understand deeply — it answers two questions that every lender must answer before extending credit, and no one can replicate the data required to answer them. The credit bureau oligopoly is one of the most durable competitive structures I have observed in my career, and the Twin verification database adds a layer of differentiation within the oligopoly that is genuinely unique. No amount of capital can accelerate the replication of 200 million active employment records built through individual employer integrations over two decades. The business earns its revenue through mandatory toll collections on credit decisions — lenders cannot legally avoid credit checks, and mortgage lenders cannot efficiently avoid income verification. This is the type of inevitable, essential-service business I look for.</p><p>However, I am troubled by the 8% ROIC and the 820 basis point compression in operating margins since 2016. Revenue has nearly doubled while net income has grown only 34% — meaning every dollar of new revenue has produced only 17 cents of bottom-line profit versus 30 cents historically. The contrarian analysis correctly identified this as the central tension: is this a temporary investment trough or a permanent structural impairment? Three years of margin stagnation at 17.7-18.5% despite cloud completion being nearly finished gives me pause. I need to see the math work — D&A must decline, margins must expand, and the capital base must stabilize before I am confident that this franchise can earn the returns its structural position deserves.</p><p>I would begin buying at $145, which provides a roughly 22% margin of safety from my $185 fair value estimate. At that price, even if margins never recover beyond 20%, I earn a reasonable return from a durable oligopoly with a widening verification moat. I would not buy aggressively until GAAP operating margins cross 20% for two consecutive quarters — that is the evidence that the harvest phase has actually begun.

Let me invert this. How do I lose money owning a natural oligopoly where the last new entrant arrived 50 years ago? Three ways: permanent margin impairment from the cloud transformation that never pays off, antitrust or regulatory action against the Twin monopoly, or management continuing to destroy value through overpriced acquisitions. The first risk is real — three years of flatlined margins despite 90% cloud completion is troubling. The second is a tail risk but genuine — the CFPB's three Civil Investigative Demands and the antitrust lawsuit against Workforce Solutions directly threaten the company's most valuable asset. The third appears to be abating — the FY2025 pivot to $928M in buybacks at depressed prices suggests management has finally learned that returning cash beats acquiring businesses at premium multiples.</p><p>What I like is the simplicity of the business model. A smart 12-year-old could understand how Equifax makes money: they collect data about whether people pay their bills and whether they have jobs, and they sell that data to banks who need to make lending decisions. The data compounds over decades, the cost of serving each additional query is near zero, and no competitor can replicate the database. This passes my simplicity test. What concerns me is the management tendency toward promotional communication — using the word 'strong' 43 times in a single earnings call while operating margins sit 800 basis points below historical levels is the kind of disconnect that makes me uncomfortable.</p><p>At $135, the risk-reward is acceptable for a patient investor willing to wait 3-5 years for margin evidence. I would not commit significant capital above $150. And I would never pay 30x GAAP earnings for a business earning 8% on its capital — that is the kind of mistake that looks obvious in hindsight.

Equifax is a toll booth. Every mortgage in America requires a credit check and increasingly an income verification. The underlying activity — lending — will happen regardless of interest rate cycles. Volumes fluctuate, but the toll is mandatory. This is precisely the same structural advantage I see in FICO, Moody's, Visa, and MSCI. The fact that mortgage volumes cycle does not make this a cyclical business — it makes it a volume-variable monopoly with 100% take rates on activity that will inevitably occur. I own this business in my concentrated portfolio of 8-10 positions because the Twin database is one of the most remarkable data assets in the American economy — 200+ million active records growing at 11% annually with near-zero marginal cost per incremental query.</p><p>The 8% ROIC is the temporary consequence of a massive investment cycle, not evidence of structural impairment. I have seen this pattern repeatedly in my 18 years of venture capital experience — a business invests heavily to upgrade its infrastructure, ROIC temporarily declines as the capital base inflates, and then operating leverage kicks in as the investment matures and CapEx normalizes. The cloud transformation spent $3+ billion building infrastructure that can now support substantially more revenue without proportional cost increases. CapEx has already declined from $625M to $481M and will continue falling. The D&A charge of $714M must mathematically decline over 3-5 years as cloud assets depreciate. This is not speculation — it is accounting certainty.</p><p>I would add to my position at $155 and continue accumulating below that level. The government verification opportunity — $5 billion TAM for social services with SNAP continuous evaluation contracts already signed — represents a structural expansion of the toll booth into a new domain where Twin has no competitor. This is the type of multi-decade compounding opportunity that I build positions around over years. Companies take many years to get off the rails when the product is essential, and Equifax's products are as essential as any in the financial infrastructure ecosystem.

Who is forced to sell? Growth investors who bought EFX at $280-300 expecting 15%+ revenue growth are dumping the stock as growth has decelerated to 7%. The five-year total return of 112% versus the S&P 500's 182% has triggered style-box migration — relative-performance-sensitive funds are rotating out of what they see as a growth disappointment. But the business has not deteriorated — it has actually improved. FCF went from negative in 2021-2022 to $1.13B in 2025. The cloud transformation is 90% complete. Twin's records grew 11%. Management bought back $500M in stock at depressed Q4 prices. Everything operationally is moving in the right direction while the stock price moves in the wrong direction. That is my favorite setup.</p><p>The reflexivity here is favorable — the stock price decline does not cause business deterioration. Customers must still pull credit reports and verify income regardless of where EFX trades. This is a lagging price reflecting sentiment, not a doom loop where cheap equity prevents the company from operating. The macro overlay is constructive: if the Fed cuts rates in 2026, mortgage originations are the primary beneficiary, and Equifax's 20% mortgage revenue exposure becomes an earnings accelerator at near-100% incremental margins. The asymmetry is roughly 1.5-2:1 — limited downside given the oligopoly floor, meaningful upside if margin recovery plus mortgage normalization materializes.

Applying my moat framework, Equifax's moat hierarchy is multi-layered but mixed in quality. The Twin verification database is a Tier 1 cost savings moat — it saves lenders 3-5x versus manual verification at near-zero marginal cost — and a Tier 1 network effect moat — each new employer increases hit rates for all lender customers. These are the most customer-aligned moat types, and they are actively widening. The credit bureau oligopoly provides Tier 2 switching cost protection and Tier 3 regulatory barrier protection — less aligned with customers but structurally durable. The moat trajectory is the critical question: Twin is clearly WIDENING (11% record growth, 16 new partnerships), while the credit bureau position is STABLE (shared with Experian and TransUnion with no differentiation).</p><p>My concern is management quality. I apply the sledgehammer test: would I trust Mark Begor locked in a room with the fate of my capital? The FY2025 buyback pivot at depressed prices is encouraging, but a decade of ROIC-dilutive acquisitions totaling $5.8 billion gives me pause. The Appriss Insights acquisition at $2.9 billion — funded entirely by debt — has not yet demonstrated returns justifying the capital deployed. The 39% GAAP-to-adjusted earnings gap and the promotional earnings call tone suggest a CEO more comfortable managing narrative than delivering operational excellence. This is not a founder-led business, and it does not pass my sledgehammer test with high confidence.</p><p>I can reach my 15% hurdle rate only at $140 or below, using conservative margin recovery assumptions and a 15% discount rate. This is a good business at the wrong price for my framework. I would become a buyer if the stock drops another 15-20% or if operating margins visibly cross 20%, confirming the harvest thesis. The moat is widening where it matters most — in verification — and that gives me confidence the business will be more valuable in 5 years than today. The question is whether the current price already reflects that.

Interesting business. The Twin database is a genuinely unique asset, and the oligopoly structure is one of the most durable I have observed. No new credit bureau has been created in 50 years — that tells you everything about the barriers to entry. The verification monopoly within the oligopoly adds a layer of differentiation that the other two bureaus cannot match. If I were building a portfolio from scratch, this is the type of industry structure I would want exposure to.</p><p>But I am not a growth investor. At $169.50 and $20.8B market cap, the size gate is clear — this is well within my range. My concern is the P/E. At 31x GAAP earnings, this exceeds my 20x threshold. Even using normalized EPS of $7.50 (assuming margin recovery to 22%), the P/E drops to 22.6x — still above my ceiling. I need to see this in the low-to-mid teens P/E on normalized earnings to get excited. That means either a price decline to $120-130 or a substantial earnings improvement that brings trailing P/E below 20x. The math simply does not work at current prices for an asymmetric bet.</p><p>I would also want to clone the position from someone I trust. Bill Nygren at Oakmark holds a 3.5% allocation — his involvement is a positive signal. But I would want to see him adding at current levels, not just holding a legacy position. If the stock drops to $120-130, the P/E approaches my threshold, the asymmetry improves to 2:1+, and I would seriously consider a starter position.

My evolutionary lens asks one question: has this business demonstrated survival fitness across multiple crises? Equifax's answer is mixed. The 2017 data breach was an existential test — 147 million Americans' data exposed, stock collapsed, operating income turned negative in 2019, and the company spent $3+ billion rebuilding its technology infrastructure. The business survived and revenue has nearly doubled since. That is evidence of adaptive capability. The Twin database — which was the strategic pivot that saved the franchise — represents evolutionary innovation born from crisis, much like how organisms develop new survival mechanisms under environmental pressure.</p><p>However, the evolutionary fitness is incomplete. Operating margins of 18% versus pre-crisis 26% suggest the organism has survived but not fully recovered. The metabolic efficiency — ROIC of 8% — is the lowest in 14 years. A truly fit organism would have emerged from crisis with improved economics, not merely larger revenue on worse margins. I need to see the metabolic rate improve before I can confirm evolutionary fitness. The industry environment is favorable — slow-changing, oligopolistic, with barriers measured in decades not years. This is exactly the type of environment where adapted survivors compound.</p><p>I would begin accumulating at $140, where the margin of safety is adequate for the remaining uncertainty around margin recovery. The business will almost certainly exist in 20 years — the credit bureau oligopoly is one of the most durable structures in the economy. The question is whether it will earn excellent returns or merely adequate ones. At $140, I am adequately compensated for that uncertainty.