Análisis profundo
Desglose paso a paso
Paso 1. Why the Same Collection Produces Different Scores
A single collection account on your credit report can produce score swings of 50-100+ points depending on which scoring model is being used. This is not a bug -- it is a fundamental design difference between scoring generations. FICO and VantageScore have each released multiple versions, and each version treats collections differently. The problem is that consumers almost never know which model their lender is using at the point of application.
FICO alone has over 50 active scoring models in the market. FICO 8 (released 2009) is still the most widely used for general-purpose credit decisions. FICO 9 (released 2014) introduced major changes to collection handling. FICO 10 and FICO 10T (released 2020) added trended data. Each of these treats paid collections differently, and each is used by different lender segments.
VantageScore, developed jointly by Equifax, Experian, and TransUnion, has its own generational differences. VantageScore 3.0 (2013) and VantageScore 4.0 (2017) both ignore paid collections, but they differ in how they weight unpaid collections by age and amount. The result is that a consumer's score can vary by 80+ points across models -- all using the same underlying credit report data.
- FICO has 50+ active scoring models -- the version your lender uses determines the collection impact
- Score differences of 80-100+ points across models are common for consumers with collections
- VantageScore models are developed jointly by the three major credit bureaus
- Consumers rarely know which scoring model a lender uses at the point of application
Paso 2. FICO 8: Paid and Unpaid Collections Are Treated the Same
FICO 8 is the critical model to understand because it remains the dominant scoring version for credit card approvals, auto lending, and many personal loan decisions. Under FICO 8, a paid collection and an unpaid collection with an original balance above $100 carry the same negative weight. The algorithm treats the existence of the collection tradeline as the derogatory event -- not whether you subsequently paid it.
This design choice reflects FICO's view that historical payment behavior is predictive of future behavior. The fact that you had an account go to collections is, in FICO 8's model, a risk signal regardless of whether you later resolved it. The scoring penalty from a collection under FICO 8 typically ranges from 50-110 points depending on the consumer's overall file thickness, with thin files seeing the largest impact.
FICO 8 does have one important exception: it ignores collections with an original balance under $100. This carve-out was designed to prevent small nuisance debts -- a forgotten library fine or parking ticket -- from causing disproportionate score damage. However, this threshold has not been adjusted for inflation since 2009, and many consumer advocates have argued it should be raised to $500.
- Under FICO 8, paid collections above $100 carry the SAME penalty as unpaid collections
- FICO 8 ignores collections with original balances under $100
- Typical FICO 8 score impact from a collection: 50-110 points depending on file thickness
- FICO 8 is still the dominant model for credit cards, auto loans, and personal loans as of 2025
Paso 3. FICO 9 and FICO 10: The Paid Collection Distinction
FICO 9, released in 2014, was the first FICO model to differentiate between paid and unpaid collections. Under FICO 9, a collection that has been paid in full is excluded from the scoring calculation entirely -- effectively treated as if it does not exist. An unpaid collection still carries the same penalty as under FICO 8. This was a major policy shift that aligned FICO with VantageScore's approach.
FICO 10 (standard) maintains FICO 9's treatment of paid collections. FICO 10T, the trended data variant, goes further by analyzing 24 months of payment behavior patterns. Under FICO 10T, a consumer who paid a collection and has shown improving credit behavior across multiple tradelines can see a larger score recovery than under FICO 10 standard, because the trended data captures the positive trajectory.
The adoption gap between FICO 8 and FICO 9/10 is the core problem. Despite being available for over a decade, FICO 9 has seen limited adoption among lenders. Most credit card issuers, auto lenders, and personal loan providers still use FICO 8 or industry-specific FICO variants (like FICO Auto Score 8 or FICO Bankcard Score 8). The FHFA mandate for FICO 10T in conforming mortgages starting Q4 2025 is the largest single adoption event for newer FICO models.
- FICO 9: paid collections are excluded from scoring entirely; unpaid collections still penalize
- FICO 10 maintains FICO 9's paid collection treatment; FICO 10T adds 24-month trended data
- FICO 9 adoption remains limited despite being available since 2014
- FHFA mandated FICO 10T for conforming mortgages starting Q4 2025 -- the largest adoption event
Paso 4. VantageScore 3.0 and 4.0: Ignoring Paid Collections Since 2013
VantageScore 3.0, released in 2013, was the first major scoring model to completely exclude paid collections from its calculations. VantageScore 4.0, released in 2017, maintained this approach and added machine learning components that further reduced the weight of unpaid collections based on age and amount. Both models also exclude medical collections regardless of payment status -- a policy the bureaus later adopted in their own voluntary reporting changes.
VantageScore's market share has grown significantly but remains secondary to FICO for most lending decisions. As of 2025, VantageScore is estimated to be used in roughly 20-25% of credit decisions, with strongest adoption in the fintech lending segment (Upstart, LendingClub, SoFi), personal loan origination, and tenant screening. Major credit card issuers and auto lenders still predominantly use FICO variants.
One area where VantageScore dominance matters for consumers: the free credit scores provided by Credit Karma, Chase Credit Journey, and most banking app score features use VantageScore 3.0. This means the score a consumer sees in their banking app will already reflect the paid-collection exclusion, even if the lender they apply to uses FICO 8 and sees a significantly lower score. This 'score gap' is one of the most common sources of consumer confusion.
- VantageScore 3.0 (2013) and 4.0 (2017) both exclude paid collections entirely from scoring
- VantageScore is used in ~20-25% of credit decisions, with strongest adoption in fintech lending
- Credit Karma and most free banking app scores use VantageScore 3.0, not FICO
- The gap between VantageScore (which ignores paid collections) and FICO 8 (which does not) causes widespread consumer confusion
Paso 5. Lender-Specific Model Usage: Who Uses What
Mortgage lenders have been the most standardized segment. Until Q4 2025, conforming mortgages required Classic FICO scores (versions 2, 4, and 5 -- yes, even older than FICO 8). Under these legacy models, paid collections still carry weight. The FHFA transition to FICO 10T and VantageScore 4.0 will change this, but the transition timeline extends through 2026 as servicers and technology vendors update their systems.
Auto lenders predominantly use FICO Auto Score variants, which are industry-specific models calibrated for auto loan default prediction. FICO Auto Score 8 treats collections similarly to base FICO 8 -- paid collections still penalize. However, auto lenders have wider manual underwriting flexibility than mortgage lenders, so a loan officer can often override a score-based decline if the collection is paid and the rest of the file is clean.
Credit card issuers are the most varied in model usage. Capital One uses VantageScore for some pre-qualification decisions but FICO for actual underwriting. American Express uses FICO Bankcard Score 8. Discover has historically used TransUnion FICO 8. Chase uses Experian FICO 8 for most card applications. This fragmentation means the same consumer with the same collection can be approved by one issuer and declined by another purely based on model differences.
- Mortgage: transitioning from Classic FICO (2/4/5) to FICO 10T and VantageScore 4.0 by 2026
- Auto: predominantly FICO Auto Score 8, which treats paid collections similarly to base FICO 8
- Credit cards: varies by issuer -- Capital One uses VantageScore for pre-qual, Chase uses Experian FICO 8
- Fintech lenders (Upstart, SoFi, LendingClub): predominantly VantageScore models
Paso 6. The Strategic Decision: When Paying a Collection Makes Sense
Given the model fragmentation, paying a collection makes the most financial sense when the consumer's target lender uses FICO 9, FICO 10, or VantageScore -- because payment alone removes the scoring penalty. For consumers targeting FICO 8 lenders, payment without deletion produces no score improvement, and the strategic calculus shifts toward negotiating pay-for-delete or focusing resources on other score factors.
There is a moral hazard argument here that the industry debates internally. If paying a collection does not improve your FICO 8 score, why pay? The answers go beyond scoring: unpaid collections can be sued upon within the statute of limitations (typically 3-6 years depending on state), collectors can continue calling and reporting, and some lenders manually review credit reports and view paid collections more favorably than unpaid ones even if the score does not change.
For consumers with multiple collections, the order of payment matters under FICO 9/10 and VantageScore. Pay the largest balance first, because larger collections contribute more negative weight in these models even when they exclude paid items. Also consider the age of each collection -- items approaching the 7-year FCRA reporting limit will fall off naturally, so paying a collection that drops off in 8 months provides less incremental value than paying one with 4 years remaining.
- Payment without deletion helps under FICO 9/10 and VantageScore but does nothing under FICO 8
- Unpaid collections carry legal exposure -- collectors can sue within state statutes of limitation (typically 3-6 years)
- Manual underwriting can favor paid collections even when scores do not change
- Pay largest balances first; skip collections within 8-12 months of the 7-year reporting limit
