Most people think regulatory pressure forces monopolies to retreat. FICO did the opposite.
In October 2025, the company unveiled a new direct-licensing model that could generate $113.85 in royalties per mortgage—a 667% increase over the $14.85 it collected historically—and in the same stroke made the credit bureaus' markup completely transparent, shifting political heat while tightening the rent-extraction ratchet even further. The move wasn't a concession. It was escalation disguised as compliance, and it revealed something crucial about how monopoly rent extraction works in its terminal phase: FICO can keep raising prices precisely because it hasn't innovated in forty years.
The mechanics are simple but brutal. Since 2018, FICO has raised the price of a single credit score from $0.60 to $4.95—a 725% increase. The new direct-licensing model offers lenders two options: pay the $4.95 flat fee plus a $33 funded-loan fee when the mortgage closes, or pay a flat $10 per score with no additional fees. On a tri-merge basis—three scores pulled from Equifax, Experian, and TransUnion—that first option delivers $113.85 to FICO versus the historical $14.85. The second option doubles revenue to $30 per tri-merge. Both represent massive acceleration on top of the 725% already extracted, and both were framed as transparency measures to expose bureau markup. The FHFA had blamed FICO for rising mortgage costs; FICO responded by making it crystal clear that bureaus were marking up the score by as much as 1,844% before the price hikes began. The regulator wanted a villain.
FICO can keep raising prices by 667% on top of 725% without losing a single meaningful customer—is that switching costs aren't economic. They're systemic.
FICO pointed at the intermediaries and kept hiking.
What makes this work—what allows FICO to raise prices by 667% on top of 725% without losing a single meaningful customer—is that switching costs aren't economic. They're systemic. A bank's risk engine is hardcoded with FICO bands. A 700 FICO score is not equivalent to a 700 VantageScore, even though VantageScore uses the same 300-to-850 range. To switch, a lender would need to re-underwrite millions of past loans to recalibrate what each VantageScore band actually means in terms of default risk. Get that recalibration wrong, and the bank faces billions in unexpected defaults. The cost of switching isn't the subscription fee—it's the existential risk of mispricing credit across the entire portfolio.
The securitization market makes it worse. Mortgage-backed securities investors price pools based on average FICO scores. The to-be-announced market, where more than 90% of agency MBS trade, is standardized around FICO. A pool of loans with VantageScore credit profiles introduces heterogeneity, and heterogeneity in securitization markets gets punished with risk premiums. Estimates suggest that for every basis point of VantageScore risk premium over FICO, a borrower pays more than $900 extra over the life of a 30-year, $400,000 mortgage. That's a $900+ penalty to save $50 in FICO royalties—an 18-to-1 cost ratio that makes the monopoly unassailable. Even when Equifax offered VantageScore 4.0 at $4.50—less than half FICO's price—lenders didn't switch. VantageScore was already bundled free with FICO pulls, and still no one used it.
When free isn't cheap enough, you're not dealing with a product monopoly. You're dealing with infrastructure lock-in.
The result shows up in the financials with almost obscene clarity. FICO's Scores segment posted 88% operating margins in fiscal 2025. Incremental operating margins—the marginal profit on each additional dollar of score revenue—hit 95%. For every extra dollar FICO collects in score fees, 95 cents flows straight to operating income. The Scores segment generated $1.2 billion in revenue at those 88% margins, while the Software segment—everything else FICO does—posted $822 million at 30% margins. Even though Software represents 40% of revenue, it contributes only 20% of operating income. FICO is two businesses stapled together, but only one matters, and that one has margins that make software-as-a-service companies look like commodity retailers.
What's striking is how little of that profit gets reinvested in the product. The FICO score algorithm—the actual mathematical model that determines your three-digit number—remains fundamentally unchanged since the 1980s. It still weighs payment history and credit utilization most heavily, followed by length of credit history, mix of credit types, and recent inquiries. FICO has released updated versions—FICO 8, FICO 9, FICO 10—but the core methodology hasn't shifted. The company doesn't disclose R&D spending specific to the score business, but the overall picture is clear: capital allocation is overwhelmingly tilted toward buybacks and dividends, not innovation. Management's incentives are locked entirely on capital return, not product development, because there's no competitive pressure to develop the product. The monopoly protects the stasis, and the stasis protects the monopoly.
This is where the paradox gets uncomfortable. FICO can keep raising prices because the score is embedded in systems that would take years and billions to replace—but that embedding is a function of the score being a stable, unchanging standard. The three-digit number works as infrastructure precisely because it hasn't evolved. Banks, securitization markets, regulators, and borrowers all understand what a 700 FICO score means because it's meant roughly the same thing for four decades. Innovation would destabilize that shared understanding, creating friction in every system that depends on FICO's consistency. So FICO doesn't innovate, and the lack of innovation becomes the moat.
But that same stasis creates surface area for a different kind of disruption. The threat isn't VantageScore—a competitor offering a functionally identical three-digit score at a lower price. The threat is AI-native underwriting that bypasses the three-digit score entirely. If lenders can feed raw credit-bureau data, transaction histories, and alternative data into machine-learning models that predict default risk more accurately than a single summary score, the infrastructure lock-in dissolves. The hardcoded FICO bands become irrelevant if the risk engine isn't using bands anymore. The securitization premium disappears if MBS investors start pricing pools based on model-predicted default probabilities rather than average FICO scores.
The switching cost evaporates because there's nothing to switch—lenders would be moving to a fundamentally different underwriting paradigm, not replacing one score with another.
The regulatory scrutiny accelerates the timeline. Senator Josh Hawley sent an oversight letter to FICO in March 2026, calling FICO's pricing "an exercise of monopoly pricing power." FHFA Director Pulte blamed FICO for unsustainable housing costs, even though FICO's royalties represent 0.04% of total mortgage production expenses. Both regulators missed the point—FICO's fees are trivial relative to closing costs, and the real markup came from credit bureaus—but the political pressure creates space for alternative underwriting models. Regulators have already mandated that Fannie Mae and Freddie Mac accept non-FICO scores, and if AI underwriting models gain traction in that regulatory window, the systemic lock-in unravels faster than FICO can finish harvesting.
FICO's stock reflected the tension. The share price peaked around $2,380-2,400 in late 2024, then collapsed 45% to $1,300 by August 2025 as regulatory noise intensified. Investors priced in monopoly rents extending indefinitely, then repriced when the durability of those rents came into question. But the business kept compounding. Fiscal 2025 delivered 16% overall revenue growth, 27% Scores growth, 46% operating margins, and 45% free cash flow margins. The direct-licensing model hadn't even begun to hit the financials. If it does—if lenders adopt the $33 funded-loan fee model at scale—Scores revenue could double or triple from current levels, all at 95% incremental margins.
The bull case writes itself: FICO extracts another five to seven years of monopoly rents at accelerating prices, returns tens of billions to shareholders through buybacks, and the stock compounds at 20%-plus annually until AI underwriting becomes a viable threat. The bear case is equally clear: AI underwriting adoption happens faster than expected, regulators force the GSEs to accept alternative models, and FICO's 88% margin business evaporates before management finishes harvesting. The paradox is that both cases depend on the same fact. FICO can keep raising prices because it hasn't innovated since the 1980s, and it will remain vulnerable to disruption for exactly the same reason.
The terminal state of monopoly rent extraction isn't sustainable dominance. It's the recognition that the moat is also the liability—that the very stability and stasis that allow for 95% incremental margins create the conditions for sudden, total obsolescence. FICO is racing to extract maximum value before the three-digit score becomes a relic. Whether it finishes that race depends on how fast AI underwriting moves from theory to infrastructure, and whether regulators accelerate or delay that transition. Right now, FICO is winning. But the urgency embedded in its pricing strategy suggests management knows the clock is running.