PCAF Motor Vehicle Loans Emissions
A motor vehicle loan finances a single, identifiable physical asset that emits in direct proportion to how far it is driven and what it burns. Unlike a corporate bond or a business loan, there is no diversified company behind the borrower — the emissions object is the vehicle itself, and the financier’s claim on it is unusually clean to measure. What makes Asset Class 6 hard is not the formula; it is sourcing the four numbers the formula needs across a portfolio of ten thousand small-ticket loans.
The methodology is mechanically simple — and operationally the most data-intensive of any PCAF asset class.
PCAF Asset Class 6 attributes each financed vehicle’s annual emissions to the lender in proportion to the outstanding loan balance divided by the total vehicle value at origination, then sums across the portfolio. Each loan is scored 1–5 for data quality.
This page is the citation-grade treatment of the PCAF motor vehicle loans asset class as defined in the PCAF Global GHG Accounting and Reporting Standard for the Financial Industry. It covers the asset-class boundary against business loans and mortgages, the outstanding-amount over total-vehicle-value attribution principle, the per-vehicle annual emissions object across tailpipe and well-to-tank and electricity, the five-rung data quality score, the unique Score 1 = primary fuel-consumption rung, electric and plug-in hybrid treatment including the charging-location assumption and the utility-factor problem, three audit-record worked examples, and the assurance and disclosure obligations. For sibling asset classes see the listed equity and corporate bonds methodology, the business loans and unlisted equity methodology, the project finance methodology, and the commercial real estate methodology.
What Asset Class 6 Covers: Loans Secured Against Identified Vehicles
PCAF Motor Vehicle Loans — Asset Class 6 in the Global GHG Accounting and Reporting Standard for the Financial Industry — covers loans and credit facilities provided to retail and business customers to finance the purchase of one or more identifiable motor vehicles, where the vehicle itself is the asset against which the loan is secured. The financed object is the vehicle; the emissions to attribute are its operational annual emissions; the denominator is the vehicle’s value at the time the loan was originated. This narrows the methodology in two important ways. First, the financier is not lending against a diversified balance sheet, so there is no enterprise value to deal with. Second, the emissions follow almost mechanically from a small set of physical parameters — make, model, fuel type, annual distance — which makes the calculation easy in principle and demanding in practice because each retail loan is a small ticket and a portfolio may contain tens of thousands of them.
Inside the boundary
Asset Class 6 covers retail auto loans for personal vehicles, fleet loans for company-owned vehicles where the loan is collateralised against the specific vehicles, motorcycle and light-commercial-vehicle loans, and refinancing of any of the above. It applies to both new and used vehicles. The unifying feature is loan-vehicle pairing — the lender can point to a vehicle identification number, a make and model, an origination value, and the borrower’s intended use.
Outside the boundary
General-purpose consumer credit (a personal loan that the borrower happens to spend on a car) is not motor vehicle finance in PCAF terms — it lacks the loan-vehicle pairing. Operating leases, where the lessor retains ownership and the vehicle never appears on the customer’s balance sheet, do not belong in Asset Class 6 either; the lessor records the vehicle in its own operational inventory rather than through a financed-emissions line. Finance leases — economically a purchase financed over time — are a judgement call: where the lease transfers substantially all the risks and rewards of ownership to the lessee, the position generally behaves like a secured vehicle loan and is treated accordingly. Loans to dealerships for floorplan inventory, fleet operating leases sold as a service, and corporate revolving credit facilities used opportunistically for vehicle purchases belong elsewhere — typically in business loans and unlisted equity.
The classification test is whether the loan is secured against one or more identifiable vehicles. If the lender holds vehicle-level collateral and can match each loan to specific VINs or make-model records, it sits in Asset Class 6. If the loan is general-purpose consumer or business credit without per-vehicle collateral, it does not.
Where this sits in the financier’s own inventory
As with every PCAF asset class, motor vehicle loans roll up into the financier’s own Scope 3 Category 15 (Investments) line under the GHG Protocol Corporate Standard. The borrower’s own scope structure — tailpipe combustion as the vehicle’s Scope 1, electricity for charging as the vehicle’s Scope 2, and well-to-tank as the vehicle’s Scope 3 Category 3 — is preserved within the attributed total so that the financier can disclose the scope split. This becomes operationally important when a borrower’s portfolio mixes ICE and electric vehicles, because the dominant scope flips between Scope 1 and Scope 2 with the powertrain.
Provider directory
Now it has to stand up to your auditor.
See who does this work. Every listing names the standards it works to, and paid placements are labelled. Including Persefoni and SE Advisory Services.
Browse 3 financed emissions providers →Do this work? A listing is US$390 a year. Get listed →
The Financed-Emissions Equation
For each financed vehicle, financed emissions in a reporting year equal that vehicle’s annual operational emissions multiplied by the loan’s attribution factor:
| Financed emissions (per vehicle, per year) |
|---|
| Financed emissions = Attribution factor × Vehicle annual emissions |
| Attribution factor = Outstanding amount ÷ Total vehicle value at origination |
| Vehicle annual emissions = Annual distance × Per-km emission factor |
| = Annual fuel use × Per-unit emission factor (the Score 1 form) |
The portfolio total is the straight sum of per-vehicle financed emissions. There is no netting against zero-emission vehicles, no portfolio-level denominator, no offsetting from any other product line. Every vehicle is computed individually and added, which keeps the result traceable and assurable down to the VIN.
The attribution factor: outstanding amount ÷ vehicle value at origination
The numerator is the outstanding loan balance at the reporting date — the unpaid principal the financier still holds against the vehicle. For balloon-structured loans with a residual-value component (common in retail auto finance), the outstanding amount includes both the principal balance and the unpaid residual that the lender remains exposed to. The denominator is the total value of the vehicle at the time the loan was originated, typically the on-the-road purchase price recorded in the financing documents. The factor is dimensionless and, on a correctly classified loan, sits between 0 and 1; values above 1 are almost always an error in the origination-value record or a loan-classification mistake.
Two structural points follow. First, the denominator is fixed at origination and does not move as the vehicle depreciates — using the current market value of a five-year-old car as the denominator would inflate the attribution factor as the asset ages, which is not what the methodology intends. Second, the attribution factor declines naturally over the loan’s life as the borrower pays down principal, which is the correct economic behaviour: the financier finances a shrinking share of the vehicle over time.
Vehicle annual emissions: distance × per-km factor, or fuel × per-unit factor
The emissions object is the vehicle’s operational annual emissions in the reporting year. Two equivalent formulations are valid, and the data quality ladder is built around which one the institution can support with evidence. The distance form — annual kilometres travelled multiplied by a vehicle-specific per-km emission factor — is the dominant formulation in retail portfolios where distance is estimated from regional defaults or telematics. The fuel-consumption form — annual litres of petrol or diesel multiplied by a fuel emission factor — is the Score 1 form, available when the lender has access to the borrower’s actual fuel-card or telematics data. The two reconcile exactly when the per-km factor is decomposed into fuel-consumption rate × fuel emission factor, which is why a Score 1 fuel record beats a Score 3 distance estimate: the rate (litres per km) is the dominant source of variance across drivers.
| Dimension | Motor vehicle loans (Asset Class 6) | Listed equity / business loans |
|---|---|---|
| Financed object | One identifiable vehicle | A whole company |
| Denominator | Total vehicle value at origination | EVIC (listed) or company equity + debt (unlisted) |
| Emissions attributed | The vehicle’s annual operational emissions | The company’s whole footprint |
| Numerator | Outstanding loan balance at reporting date | Outstanding amount in the company |
| Strongest data rung in practice | Primary fuel/electricity records (Score 1) | Reported company emissions (Scores 1–2) |
| Typical portfolio shape | Tens of thousands of small-ticket loans | Hundreds of larger positions |
| Attribution factor over time | Falls as loan amortises; denominator fixed at origination | Varies with position value and denominator |
Scope Coverage — Tailpipe, Well-to-Tank, and Electricity
A vehicle’s annual emissions are not a single number; they decompose by scope on the vehicle’s own inventory, and the financier carries the same split through into the attributed line. The required scopes are the vehicle’s Scope 1 (tailpipe combustion for ICE) and Scope 2 (purchased electricity for EVs); the vehicle’s Scope 3 Category 3 (well-to-tank fuel supply-chain emissions, and for EVs, transmission and distribution losses on the electricity drawn) is phased in, with the expectation that financial institutions disclose it where material.
| Vehicle scope | What it captures | Powertrain | Treatment in the financier’s attributed total |
|---|---|---|---|
| Scope 1 (tailpipe) | Direct combustion CO₂, CH₄, N₂O emitted from the vehicle exhaust | ICE, HEV, PHEV (charge-sustaining mode) | Required |
| Scope 2 (electricity) | Emissions from grid electricity used to charge the traction battery | BEV, PHEV (charge-depleting mode) | Required |
| Scope 3 Cat 3 (WTT, fuels) | Upstream extraction, refining, and distribution of petrol/diesel | ICE, HEV, PHEV | Phased — required where material |
| Scope 3 Cat 3 (T&D, electricity) | Grid transmission and distribution losses on charging electricity | BEV, PHEV | Phased — required where material |
The methodological reference for the WTT component is the well-to-tank methodology; for the electricity component the location-based vs market-based choice is discussed in the Scope 2 electricity methodology and the market-based companion. T&D losses on the EV electricity component are covered in the T&D losses methodology. The financier follows the borrower’s scope structure — the per-km factors in DEFRA, EPA, and EEA datasets are typically decomposed into the tailpipe, WTT, and (for EVs) electricity components, and that decomposition flows directly into the attributed scope split.
Reporting only tailpipe is a material understatement. WTT typically adds 15–25% to a diesel or petrol vehicle’s per-km footprint depending on the fuel pathway, and the entire emissions of a BEV live in Scope 2 plus its T&D component — omitting Scope 2 reports a battery-electric vehicle as effectively zero-emission, which is inconsistent with the methodology. Disclose the scope split explicitly.
The PCAF Data Quality Score (1–5) for This Asset Class
The five-rung score works identically to the other asset classes in structure but is parameterised differently for vehicles. Score 1 here is unusual: it requires primary fuel or electricity consumption data for the specific financed vehicle, available almost exclusively from telematics integrations, fuel-card data, or fleet-management feeds. Scores 2 and 3 step down to known specifications with measured or estimated distance; Scores 4 and 5 collapse onto vehicle make-or-region averages. In the GreenCalculus MasterBrain these scores are stored as framework markers under reporting.pcaf.motor_vehicle_loans.score_1 through score_5 — rank values, not emission factors.
| Score | Activity data | Emission factor | What the institution holds | Audit risk |
|---|---|---|---|---|
| 1 | Primary fuel or electricity consumption — actual litres of petrol/diesel or actual kWh of charging energy for the financed vehicle | Verified fuel/electricity emission factor (DEFRA, country grid) | Telematics integration, fuel-card records, or charging-network records linked to the VIN | Low |
| 2 | Measured annual distance for the financed vehicle, with known make/model/year | Vehicle-specific per-km factor (make/model/year) | Odometer readings (e.g. annual MOT/inspection record) plus vehicle specification | Low–medium |
| 3 | Estimated annual distance (regional default), with known make/model/year | Vehicle-specific per-km factor (make/model/year) | Vehicle specification from origination + national/regional average annual distance | Medium |
| 4 | Estimated annual distance (regional default) | Vehicle make/segment average (e.g. petrol medium car average) without per-model specificity | Loan record showing fuel type and broad vehicle class only | Medium–high |
| 5 | Estimated annual distance | Country or regional vehicle average (e.g. UK passenger car average across all fuels) | Only that the loan is a vehicle loan in a given country | High |
Why Score 1 looks different here
In other PCAF asset classes Score 1 means a verified company inventory — a document the lender requests and the borrower files. In Asset Class 6 there is no company inventory; the vehicle is the inventory. Score 1 therefore requires primary fuel or electricity consumption data for the specific financed vehicle. Telematics integration with fleet customers, employer fuel-card data on commercial vehicle loans, and OEM-published charging-network data for connected EVs are the three practical routes. The pay-off is large: the rate of fuel consumed per kilometre is the dominant source of variance across drivers of the same vehicle model, so Score 1 eliminates a class of error that Scores 2–3 cannot reach. Pure retail portfolios with no telematics rarely reach Score 1; commercial fleet portfolios with on-board telematics often do.
Score 2 vs Score 3 — distance evidence
The distinction between Scores 2 and 3 is whether the financier has measured distance for the specific vehicle, or is using a regional default. Measured distance does not require continuous telematics — an annual odometer reading taken at vehicle inspection (the UK MOT, the German TÜV, the French Contrôle Technique) is sufficient to promote a loan from Score 3 to Score 2 if the lender ingests that record. Few retail lenders do this today, which is why Score 2 is rare in retail portfolios and Score 3 is the modal rung for well-documented retail books.
Scores 4 and 5 — vehicle make-or-region averages
Scores 4 and 5 are the fallback rungs when even the specific make and model are not on the loan record. Score 4 uses a vehicle make-or-segment average — a petrol medium car average across the manufacturer’s range, for example. Score 5 collapses entirely onto a country or regional vehicle average across all fuels and segments. Either is acceptable as a methodology floor, but both meaningfully overstate or understate individual loans (a sub-compact will be over-estimated by a national-average factor, an SUV under-estimated). The portfolio reports a single emissions-weighted average score, weighting each loan by its share of attributed financed emissions.
Illustrative data-quality distribution for a well-documented retail auto-loan book. The mass clusters at Score 3 — the structural consequence of retail portfolios that capture vehicle make and model at origination but rely on regional default distances. Commercial-fleet books with telematics show inverted distributions weighted toward Scores 1–2.
Activity Data — Annual Distance and Fuel Consumption
Annual distance is the activity-data input that determines whether a loan can reach Score 2. Three sources, in descending evidentiary strength.
Telematics or fuel-card records (Score 1)
Connected-vehicle telematics, fleet-management systems, and corporate fuel-card feeds give the actual litres of petrol or diesel consumed, or the actual kWh of grid electricity charged, for the specific financed vehicle. Multiplied by a verified fuel or grid emission factor, this is the Score 1 evidentiary chain. Common on commercial-fleet loans; rare on retail.
Annual odometer readings (Score 2)
Most jurisdictions require periodic roadworthiness inspections (UK MOT, German TÜV, French Contrôle Technique, US state inspections, Singapore inspection), and each inspection records an odometer reading. The delta between two successive readings is measured annual distance for the specific vehicle. Ingesting this into the loan record promotes a loan from Score 3 to Score 2 without touching telematics.
Regional or national default distance (Scores 3, 4, 5)
The PCAF Standard accepts regional or national average annual distance as a Score 3 fallback when measured data is unavailable. Sources include national transport statistics (UK DfT, US BTS, EU Eurostat road-transport statistics) and OEM market reports. Document the source, the year, and the segment-specificity (passenger car, light commercial) so a reviewer can replicate the figure.
If the data-collection budget is constrained, ingesting annual inspection readings for the largest-emitting loans (SUVs, light commercials, older diesels) does more for portfolio data quality than any amount of refinement to small-engine compact petrol loans. Weight the data-collection programme by emissions exposure, not by loan count.
Electric and Plug-in Hybrid Vehicles
Battery-electric and plug-in hybrid loans introduce two judgement calls that affect a portfolio’s results materially: the charging-location assumption that determines the grid factor, and (for PHEVs only) the utility factor that splits the year between charge-depleting and charge-sustaining operation.
BEV — the charging-location assumption
A battery-electric vehicle’s annual emissions are its annual kWh drawn from the grid multiplied by an electricity emission factor. The kWh come from one of three places — home (residential metering), workplace, or public charging (DC fast charging on a public network) — and PCAF does not prescribe a default mix. The three defensible defaults are (a) the country residential grid factor applied to all charging, which understates emissions where rapid public charging draws more from peaking generation; (b) a blended home/work/public mix derived from national charging-survey data; or (c) a marginal-emission factor representing the generation displaced by charging demand. Most retail-portfolio disclosures use (a) for simplicity; commercial-fleet disclosures sometimes use (b) where charging-network telematics distinguishes the energy source. Document the choice, cite the source, and apply it consistently across the BEV sub-portfolio.
Whichever grid factor is chosen, the choice between location-based and market-based reporting follows the borrower’s own Scope 2 conventions under the GHG Protocol Scope 2 Guidance. For most retail borrowers the location-based factor (the country grid average) is the only available choice; market-based requires the borrower to hold contractual instruments such as energy attribute certificates, which retail individuals rarely do.
T&D losses on the electricity component are a Scope 3 Cat 3 item on the vehicle’s inventory and propagate into the attributed total. Apply the country T&D loss rate on top of the grid factor; see the T&D losses methodology for the calculation and country-by-country loss-rate references.
PHEV — the utility-factor problem
A plug-in hybrid runs partly on grid electricity (charge-depleting mode) and partly on petrol or diesel (charge-sustaining mode). The split is the utility factor — the share of annual distance driven in charge-depleting mode. Three reference utility factors exist, and they diverge sharply:
| Utility-factor source | Charge-depleting share | Implied annual emissions | Use case |
|---|---|---|---|
| WLTP type-approval (EU) | ~70–80% | Low | Type-approval and CO₂ regulation in the EU |
| EPA combined cycle (US) | ~55–65% | Medium | US fuel-economy labels |
| Real-world observed (ICCT, Fraunhofer, JRC) | ~15–40% | High | Studies of actual PHEV driver behaviour |
Adopting the WLTP type-approval utility factor for a PHEV loan-book reports the vehicles as roughly twice as clean as the ICCT real-world evidence suggests. PCAF does not prescribe which to use, but the choice should be documented, defended against published evidence, and consistent across the PHEV sub-portfolio. Conservative practice for European portfolios is to use the WLTP factor for Scope 1 reporting (matching the legal type-approval basis) and to disclose a sensitivity using the ICCT real-world factor in the methodology note. Document the choice as a methodological judgement, not as data.
The PHEV utility factor and the BEV charging-location assumption together can move a mixed-powertrain loan-book’s reported emissions by 30–50% — more than any other single methodological choice in Asset Class 6. Lock both in the methodology note and the assurance documentation; do not let either drift between reporting years without disclosing the rebasing.
Worked Examples — Audit Records
Three audit-record snapshots. All values are hardcoded as audit records: they reconcile to their stated inputs and must continue to reconcile regardless of future data-layer changes. The DEFRA 2025 v1.0 tailpipe and WTT factors used below are dated to source and will not float on subsequent DEFRA releases; the same arithmetic appears in the PCAF Motor Vehicle Loans calculator once published.
Example A — single retail loan, diesel medium car, Score 3
| Example A — diesel medium car, retail loan, Score 3 (estimated distance + known make/model) | Value |
|---|---|
| Vehicle — VW Passat 2.0 TDI 2022 (diesel medium car) | — |
| Total vehicle value at origination | GBP 32,000 |
| Outstanding loan balance at reporting date | GBP 20,800 |
| Attribution factor = 20,800 ÷ 32,000 | 0.65 (65%) |
| Estimated annual distance (UK DfT average, diesel passenger car) | 12,500 km |
| Per-km factor — diesel medium car, tailpipe (DEFRA 2025 v1.0) | 0.16844 kg CO₂e/km |
| Per-km factor — diesel medium car, WTT (DEFRA 2025 v1.0) | 0.04212 kg CO₂e/km |
| Vehicle annual emissions (tailpipe) = 12,500 × 0.16844 | 2,105.5 kg CO₂e |
| Vehicle annual emissions (WTT) = 12,500 × 0.04212 | 526.5 kg CO₂e |
| Vehicle annual emissions (tailpipe + WTT) | 2,632.0 kg CO₂e |
| Financed emissions (tailpipe) = 0.65 × 2,105.5 | 1,368.6 kg CO₂e |
| Financed emissions (WTT) = 0.65 × 526.5 | 342.2 kg CO₂e |
| Financed emissions (total) | 1,710.8 kg CO₂e |
| Data quality score | 3 (estimated distance, known make/model) |
The lender attributes 1.71 tCO₂e from a single diesel medium-car loan in this reporting year — 1.37 tCO₂e of tailpipe Scope 1 and 0.34 tCO₂e of WTT Scope 3 Cat 3. The factor decomposition makes the scope split fall out automatically, and the audit chain runs from loan-system fields (origination value, current balance) through the activity assumption (UK DfT regional default distance) to the emission-factor source (DEFRA 2025 v1.0) without leaving any judgement undocumented.
Example B — fleet loan, battery-electric vehicle with telematics, Score 1
| Example B — BEV, commercial-fleet loan, Score 1 (primary electricity consumption) | Value |
|---|---|
| Vehicle — Tesla Model Y Long Range AWD 2024 (BEV) | — |
| Total vehicle value at origination | GBP 48,000 |
| Outstanding loan balance at reporting date | GBP 36,000 |
| Attribution factor = 36,000 ÷ 48,000 | 0.75 (75%) |
| Measured annual charging energy (fleet telematics, reporting year) | 4,200 kWh |
| Charging-location assumption | Country residential grid (GB) |
| UK grid factor — location-based (DEFRA 2025 v1.0) | 0.20493 kg CO₂e/kWh |
| UK T&D losses (% of power output, World Bank 5-yr avg) | 8.79% |
| T&D-loss factor on charged kWh = 0.20493 × 0.0879 | 0.01802 kg CO₂e/kWh |
| Vehicle annual emissions (Scope 2 generation) = 4,200 × 0.20493 | 860.7 kg CO₂e |
| Vehicle annual emissions (Scope 3 Cat 3 T&D) = 4,200 × 0.01802 | 75.7 kg CO₂e |
| Vehicle annual emissions (total) | 936.4 kg CO₂e |
| Financed emissions (Scope 2) = 0.75 × 860.7 | 645.5 kg CO₂e |
| Financed emissions (Scope 3 Cat 3) = 0.75 × 75.7 | 56.8 kg CO₂e |
| Financed emissions (total) | 702.3 kg CO₂e |
| Data quality score | 1 (primary electricity consumption) |
The BEV draws its entire footprint from Scope 2 plus the T&D component on the vehicle’s Scope 3 Category 3 — there is no tailpipe term. The Score 1 evidence chain runs from the fleet telematics directly: measured kWh × verified grid factor × T&D loss rate, with the attribution applied at the loan level. Note the attribution factor of 0.75 reflects that the loan is still 75% outstanding; over the loan’s life this will decline toward zero as the borrower repays principal, naturally reducing the attributed emissions even if the vehicle’s energy use stays constant.
Example C — mixed portfolio of five loans, weighted DQS roll-up
| Loan | Vehicle | Outstanding | Origination value | Attr. factor | Vehicle annual t CO₂e | Financed t CO₂e | Score |
|---|---|---|---|---|---|---|---|
| L1 | Diesel medium car (Passat) | £20,800 | £32,000 | 0.65 | 2.632 | 1.711 | 3 |
| L2 | BEV (Model Y) — telematics | £36,000 | £48,000 | 0.75 | 0.936 | 0.702 | 1 |
| L3 | Petrol compact (Golf 1.5 TSI) | £14,400 | £24,000 | 0.60 | 2.213 | 1.328 | 3 |
| L4 | PHEV (Octavia iV, WLTP utility factor) | £22,500 | £30,000 | 0.75 | 1.184 | 0.888 | 3 |
| L5 | Light commercial diesel van (Transit) | £28,000 | £35,000 | 0.80 | 4.890 | 3.912 | 4 |
| Portfolio total | — | £121,700 | £169,000 | — | — | 8.541 | — |
The portfolio total is the straight sum: 1.711 + 0.702 + 1.328 + 0.888 + 3.912 = 8.541 tCO₂e attributed in the reporting year. Per-vehicle annual emissions for L3 (petrol compact, 12,000 km × 0.18441 kg CO₂e/km tailpipe + WTT 0.04408 kg CO₂e/km), L4 (PHEV, WLTP-blended 0.0987 kg CO₂e/km tailpipe + electricity-charged portion 0.0247 kg CO₂e/km at 12,000 km), and L5 (light commercial diesel, 18,000 km × 0.271 kg CO₂e/km tailpipe + WTT) are computed identically to Examples A and B against DEFRA 2025 v1.0 — abbreviated in the table above to keep the portfolio view legible.
| Emissions-weighted data quality score | Value |
|---|---|
| Weighting = each loan’s financed emissions ÷ portfolio financed emissions | 1.711 / 0.702 / 1.328 / 0.888 / 3.912 of 8.541 |
| Weighted score = (1.711×3 + 0.702×1 + 1.328×3 + 0.888×3 + 3.912×4) ÷ 8.541 | (5.133 + 0.702 + 3.984 + 2.664 + 15.648) ÷ 8.541 |
| Weighted data quality score | 3.29 |
The light commercial van (L5) at Score 4 dominates the portfolio total and pulls the weighted score above 3, even though the four passenger loans (L1–L4) cluster at Scores 1 and 3. The lesson is the same as in the project-finance methodology: data-quality effort should follow the emissions, and obtaining a vehicle-make-and-model record for the largest-emitting commercial loans (promoting them from Score 4 to Score 3) does more for the portfolio’s disclosed quality than refining any of the smaller passenger loans.
Weight the data-collection programme by attributed emissions, not by loan count. A retail auto book of 50,000 loans typically has 5–10% of vehicles producing 25–35% of the attributed emissions — light commercial vehicles, older diesels, and large SUVs. Promoting those loans from Score 4 to Score 3 (or Score 3 to Score 2 via inspection-reading ingestion) compounds across the portfolio. Spending the same effort on a 1.0-litre compact petrol moves the disclosure barely at all.
Edge Cases and Known Pitfalls
Denominator is fixed at origination — not market value
The total vehicle value in the attribution-factor denominator is the value at the time the loan was originated, not the current market value of the vehicle today. Re-denominating to current market value as the vehicle depreciates inflates the attribution factor year on year (since depreciation outpaces principal repayment for the first several years), which overstates attributed emissions. The denominator is locked at origination; only the numerator updates.
Balloon and residual-value structures
Retail auto finance frequently bundles a balloon payment or guaranteed minimum future value at the end of the loan term. The outstanding amount includes the principal balance and the unpaid balloon/residual exposure the lender remains on hook for. Truncating to principal-only understates the financier’s claim and the attributed emissions; include the full unpaid balance the lender is exposed to.
Refinancing a used vehicle
When an existing loan is refinanced — by the original lender or a new one — the denominator resets to the vehicle’s value at the refinancing date, not the original purchase price. This is correct: the refinancing is a new loan secured against the vehicle’s value at that point. Document the refinancing event and the new origination value, and avoid mixing pre- and post-refinancing denominators within one year’s attribution.
Repossession or write-off mid-year
If the vehicle is repossessed or the loan is written off partway through the reporting year, the financier’s exposure ends at that date. Pro-rate the attributed emissions to the portion of the year the loan was outstanding, and document the repossession or write-off in the audit trail. A common error is carrying a written-off loan’s attributed emissions for the full year — overstates the inventory.
Securitised auto-loan pools
When an originator securitises an auto-loan pool, the financed emissions follow the credit exposure. Loans retained on the originator’s balance sheet (skin-in-the-game tranches, mezzanine retained, residual interests) remain in the originator’s Asset Class 6 inventory; loans sold to investors transfer the attributed emissions to the note-holders in proportion to their tranche exposure. The SPV itself does not carry the emissions — it is a pass-through. Allocate by where the credit risk sits at the reporting date.
Operating lease vs finance lease vs vehicle loan
Operating leases (the lessor retains the vehicle and the customer pays for use) are not Asset Class 6 — the lessor records the vehicle in its operational inventory. Finance leases that transfer substantially all the risks and rewards of ownership behave like secured vehicle loans and are treated here. Floorplan and inventory financing to dealerships is general-purpose business credit and belongs in business loans. Classify by economic substance, not by the contract’s label.
Sub-threshold high-volume retail books
A retail portfolio of tens of thousands of small-ticket loans cannot practicably be assessed loan-by-loan with vehicle-specific factors. Cohort-based estimation — grouping loans by vehicle segment, fuel type, and region, applying the same per-km factor and regional default distance to each cohort, and computing the attribution at the cohort level — is an accepted Score 3 or Score 4 approach provided the cohort definitions and weights are documented. Sample a stratified subset annually to validate the cohort assumptions against odometer or telematics evidence.
Attribution factor exceeding 1
A factor above 1 means the outstanding loan balance exceeds the vehicle’s origination value — almost always a data error (a depreciation-tracking valuation accidentally substituted for the origination value, or a loan that has been topped up post-origination without a corresponding denominator update). Cap the disclosed factor at 1, document the cap, and investigate the origination-value record before the next reporting period.
Governance, Assurance, and Disclosure Obligations
Motor vehicle loans financed emissions feed the same disclosures as the other asset classes: CSRD ESRS E1 for in-scope EU institutions, IFRS S2 for ISSB-aligned filers, the SBTi Corporate Net-Zero Standard for financial-institution target setting, the CDP climate questionnaire for voluntary disclosure, and TCFD-aligned reporting where still applicable. Asset Class 6 attracts disclosure attention because retail auto books are a visible source of consumer-facing emissions exposure and because the data quality story is unusually concrete: a regulator or assurance reviewer can ask exactly what evidence chain supports each loan’s score.
Run the following pre-flight before submitting a motor-vehicle-loans financed-emissions total for reasonable assurance:
- Loan classification. Confirm every position in scope is a loan or finance lease secured against an identifiable vehicle; exclude general-purpose consumer credit, operating leases, dealer floorplan finance, and unsecured business loans that happen to fund vehicles.
- Attribution-factor reconciliation. Every loan’s attribution factor traces to a sourced outstanding amount (drawn balance plus residual exposure where applicable) and a sourced total vehicle value at origination. Denominator locked at origination, not market-value re-denominated.
- Scope completeness. Tailpipe (vehicle Scope 1) and electricity (vehicle Scope 2) are required; WTT and T&D (vehicle Scope 3 Cat 3) are included where material, with the omission rationale documented if not.
- Per-loan data quality score. Every loan carries a 1–5 score with the evidence behind it. Score 1 is supported by primary fuel or electricity consumption records linked to the VIN; Score 2 by measured-distance evidence (e.g. inspection odometer readings); Scores 3–5 by the documented activity-data default and emission-factor source.
- EV charging-location assumption. The grid factor and charging-mix assumption for BEVs and PHEV charge-depleting mode is documented, sourced, and consistent across the EV sub-portfolio.
- PHEV utility factor. The utility factor source (WLTP, EPA, or real-world reference) is named, documented, and consistent across the PHEV sub-portfolio. A sensitivity using real-world evidence is disclosed in the methodology note.
- Emission-factor source and vintage. Per-km, per-litre, and per-kWh factors trace to a named release (DEFRA, EPA, EEA, country grid factor publication) with the version and publication year stated. Worked-example audit records hardcode the factor; future-year disclosures pick up the next release.
- Activity-data source. Annual-distance defaults trace to a national statistics source with year and segment-specificity stated. Measured-distance loans cite the odometer-reading source (inspection record, telematics integration).
- Securitisation allocation. Where the institution originates and securitises auto-loan pools, the attributed emissions follow the credit exposure at the reporting date; retained tranches stay in the originator’s inventory and sold tranches transfer to note-holders.
- Repossession and write-off pro-rating. Loans that ended mid-year are pro-rated to the months they were outstanding; no full-year attribution for written-off positions.
- Emissions-weighted DQS. The portfolio reports a single emissions-weighted average data quality score, with the calculation visible to the reviewer.
- No netting. The portfolio total is a straight sum of per-loan financed emissions; no offsetting of high-emitting loans against EV positions, no portfolio-level denominator, no avoided-emissions claim against the inventory.
The three assurance findings most specific to this asset class are market-value re-denomination of the attribution factor (denominator drifting upward year on year), PHEV utility-factor shopping (silently moving from a real-world reference to WLTP to lower reported emissions), and BEV scope-omission (reporting EVs at zero by leaving Scope 2 out). Lock all three in the methodology note: denominator at origination, utility-factor source named and stable, Scope 2 required for every EV.
What the Calculator Handles vs What You Decide
A calculator automates the mechanical attribution and roll-up arithmetic. It cannot make the classification and data-sourcing judgements that determine whether the result is assurance-grade.
The calculator handles
Computing the attribution factor (outstanding amount ÷ origination value), the per-vehicle annual emissions by scope (tailpipe + WTT for ICE; electricity + T&D for BEV; both terms for PHEV), the per-loan financed emissions, the portfolio sum, the emissions-weighted data quality score, the attribution-factor cap at 1, mid-year repossession and write-off pro-rating, the WTT and T&D adders against the right factor sources, and per-loan and per-portfolio audit trails showing the full arithmetic chain.
You decide
Whether each position is genuinely Asset Class 6 (vehicle-secured loan or finance lease) or belongs elsewhere; the data quality score assignment (Score 1 requires evidence the calculator cannot fetch); the BEV charging-location assumption and the PHEV utility-factor source; the activity-data default for non-telematics loans; the cohort definitions for high-volume retail books; and the boundary between operational and out-of-scope positions (operating leases, dealer floorplan, general-purpose credit).
Apply the vehicle-value attribution denominator, the per-vehicle annual emissions by scope, the EV charging and PHEV utility-factor judgements, and the emissions-weighted data quality score to a motor-vehicle-loan portfolio with a full per-loan audit trail.
Standards Alignment
Seven standards govern this asset class and its disclosure consequences.
| Standard | Role for this methodology |
|---|---|
| PCAF Financed Emissions Standard | Defines Asset Class 6, the outstanding-amount ÷ vehicle-value attribution formula, the per-vehicle annual emissions object across tailpipe/WTT/electricity, and the five-rung data quality score. The methodological foundation of every figure on this page. |
| GHG Protocol Corporate Standard | Defines the financier’s own inventory boundary; motor vehicle loans roll up to Scope 3 Category 15 (Investments). PCAF operationalises this category for the lender’s disclosure. |
| GHG Protocol Scope 3 Standard | Governs the materiality and phase-in logic for the vehicle’s own Scope 3 components — WTT for fuels, T&D for electricity — that propagate into the attributed total. |
| GHG Protocol Scope 2 Guidance | Governs the location-based vs market-based reporting choice for the electricity component on BEVs and PHEV charge-depleting mode. |
| CSRD ESRS E1 | EU mandatory climate disclosure; in-scope financial institutions report financed emissions and the underlying methodology, including data-quality scoring and scope breakdown. |
| UK DEFRA Emission Factors | Source for the per-km tailpipe and WTT factors used in UK-domiciled portfolios; the worked examples on this page use DEFRA 2025 v1.0 as a dated snapshot. |
| Spend-based MRIO emission factors | Supplies the Score 5 region-average fallback factors used when neither make-and-model nor segment data is on the loan record. |
Across the asset classes. This is one of ten PCAF methods. For which class applies to a given exposure, and how the attribution denominator changes between them while the equation stays the same, see PCAF Financed Emissions.
Frequently Asked Questions
The attribution factor is the outstanding loan balance at the reporting date divided by the total vehicle value at origination. For a £20,800 outstanding balance on a vehicle originally valued at £32,000, the factor is 0.65, so 65% of the vehicle’s annual emissions are attributed to the lender. The denominator is fixed at the origination date and does not move with vehicle depreciation.
Vehicles depreciate faster than loans amortise in the first few years. Re-denominating to current market value would inflate the attribution factor year on year, overstating attributed emissions and producing a perverse incentive to mark the vehicle conservatively. The methodology locks the denominator at origination — only the numerator (outstanding balance) updates over the loan’s life.
Primary fuel or electricity consumption data for the specific financed vehicle — actual litres of petrol or diesel consumed, or actual kWh charged. In practice this means telematics integration, fuel-card feeds, or charging-network records linked to the VIN. Commercial-fleet loans often reach Score 1; pure retail portfolios rarely do without an explicit telematics integration with the borrower.
Where material, yes — WTT is required as part of the vehicle’s Scope 3 Category 3 and propagates into the attributed total. WTT typically adds 15–25% to a petrol or diesel vehicle’s per-km footprint, so reporting only tailpipe understates the financed emissions materially. The DEFRA, EPA, and EEA per-km factor families publish WTT and tailpipe components separately, which makes the inclusion mechanical.
A BEV’s annual emissions are its annual charging energy (kWh drawn from the grid) multiplied by an electricity emission factor and a T&D loss factor. The grid factor depends on where charging happens — home, workplace, or public — and PCAF does not prescribe a default mix. Most retail-portfolio disclosures use the country residential grid factor; commercial-fleet disclosures sometimes derive a blended mix from charging-network telematics. Document the choice and apply it consistently across the EV sub-portfolio.
A PHEV splits its annual distance between charge-depleting mode (treated as Scope 2 electricity) and charge-sustaining mode (treated as Scope 1 tailpipe). The split is the utility factor, and reference sources diverge sharply: WLTP type-approval implies 70–80% charge-depleting, EPA labels imply 55–65%, real-world ICCT and Fraunhofer studies imply 15–40%. PCAF does not prescribe a source. Document the choice, justify it against published evidence, apply it consistently across the PHEV sub-portfolio, and disclose a sensitivity using the real-world reference.
No. Under an operating lease the lessor retains ownership and records the vehicle in its own operational inventory, not as a financed-emissions line. Finance leases that transfer substantially all the risks and rewards of ownership behave like secured vehicle loans and are treated here. The classification turns on economic substance — risk transfer and recourse — not on the contract label.
Refinancing creates a new origination event. The denominator resets to the vehicle’s value at the refinancing date, not the original purchase price, because the refinancing is economically a new loan secured against the vehicle’s value at that point. Document the refinancing date and the new origination value, and avoid mixing pre- and post-refinancing denominators within one year’s attribution.
The financier’s exposure ends at the repossession or write-off date. Pro-rate the attributed emissions to the portion of the year the loan was outstanding, and document the event in the audit trail. Carrying a written-off loan’s attributed emissions for the full year is a common error and overstates the inventory.
Cohort-based estimation is accepted. Group loans by vehicle segment, fuel type, and region; apply the same per-km factor and regional default annual distance to each cohort; compute the attribution at the cohort level. Document the cohort definitions and weights, and sample a stratified subset annually against odometer or telematics evidence to validate the assumptions. This typically supports a Score 3 or Score 4 disclosure depending on the granularity of the cohorts.