Two years before this loss, Stellantis posted a €18.6 billion profit — its best ever. The swing was not caused by a market no one could see coming. The question is not what went wrong. The question is whether it was preventable.
Was Any of This Actually Governable?
Start with the numbers, because they give the answer. The underlying adjusted operating loss was just −€0.842B on revenues of €153.5B — a margin problem, not a catastrophe. The remaining €21+ billion of the net loss was a discrete stack of management-classified adjustments, each traceable to a specific decision, a specific commitment, or a specific failure to act when the evidence changed.
These were not market losses. They were the deferred cost of execution risk that was never quantified, prioritised, mitigated, or predicted.
€9.07B — Program cancellations & supplier claims: contracts entered without modelled exit costs, cancelled after commitments were fully incurred
€6.58B — Platform impairments: €30B+ committed against a single EV base case; no scenario triggers to resize when demand data diverged
€4.13B — Warranty estimate change: "deterioration in quality as a result of operational choices" — a known operational signal never converted to financial risk
€2.05B — Battery JV resizing: gigafactory capacity committed ahead of demand certainty; one plant sold to LG Energy Solution for $100
€1.09B — Fuel cell program discontinuation: technology investment continued past the point readiness data justified it
€2.46B — Restructuring and other charges including €1.3B workforce reduction in Europe
€25.4B total. Net revenues: €153.5B (−2% YoY). Industrial free cash flow: −€4.5B.
Every charge had precursors. The warranty problem had leading quality indicators. The supplier claims had contract exposure that could have been mapped. The platform impairments had EV adoption data — publicly available — diverging from plan for multiple consecutive quarters. The workforce restructuring had hiring that was tracking a strategy narrative rather than validated market signals.
So: yes. The majority of this was governable. Not all — you cannot instrument away a macro EV demand slowdown or tariff regime change. But the €25.4B in adjustments was not the market. It was the cost of making €30B+ in irreversible commitments without a system to know when to stop.
"The underlying operating loss was −€0.842B. The other €21+ billion was adjustments. Those were not market losses. They were the deferred cost of execution risk that was never quantified, prioritised, mitigated, or predicted."
What is Really the Root Cause?
Worth answering directly. The Stellantis loss has four competing root cause explanations, each legitimate. Pinning it on "trust" alone would be intellectually lazy — and would weaken the case for what NAVETRA™ actually does.
All four are valid. And all four produce the same observable symptom: bad news doesn't travel fast enough to matter. Quality deterioration was known internally. The board and CEO held divergent views never resolved into a decision. Seven consecutive years of US sales decline were processed as variance, never escalated as pattern.
Trust is an operating condition that determines whether any governance architecture actually functions. Stellantis had ERM frameworks, board committees, quarterly reviews, and KRIs. What it lacked was the instrumentation to verify whether those structures were working — whether bad news was travelling, whether decisions reflected the information that actually existed.
NAVETRA™ is not another governance layer. It is the system that tells you whether your governance is real or performative — and converts the answer into financial terms your board can act on.
The Four Capabilities That Would Have Changed the Outcome
Strip away the theory. NAVETRA™ does four things that were absent at Stellantis at every stage of the loss. Each maps directly to specific charges in the stack.
Convert operational signals into financial exposure — in numbers, with owners, at board-visible level — before the accounting forces recognition.
At Stellantis: early-life failure rates, field incident velocity, and rework hours per unit existed as operational data. They were never translated into a warranty cost curve or Operating Income at Risk. The €4.13B provision reset is what happens when that conversion doesn't occur.
Rank execution risks by EBITDA exposure — so decision-makers act on what matters most, in sequence, before the window to act closes.
At Stellantis: seven US sales declines, deteriorating quality, and slowing EV adoption data competed for attention without a framework to force prioritisation. The €9.07B in supplier claims accumulated because contract exposure was never ranked against program risk. No one was required to decide which commitments to exit first.
Define pre-agreed response playbooks tied to evidence thresholds — so when triggers trip, the organisation pivots on data rather than crisis.
At Stellantis: "empowering regional teams to accelerate decision-making" is a post-hoc mitigation. The €2.05B battery JV resizing — including selling a plant for $100 — is what mitigation costs when it happens three years too late. Mitigation at trigger is structurally cheaper by an order of magnitude.
Model where current trajectories lead — at program, platform, and portfolio level — so impairments are management decisions, not accounting surprises.
At Stellantis: IAS 36 impairments are triggered when recoverable amount falls below carrying value — an accounting outcome of a strategic reality building for quarters. The €6.58B in platform impairments reflects a trajectory that was predictable from EV demand curves, competitor pricing, and incentive phase-outs — but was never run forward into a portfolio-level forecast with decision gates attached.
The EBITDA Case
The case for NAVETRA™ is not primarily about governance quality or trust culture — though both matter. It is about protecting EBITDA from unnecessary exposure.
"Unnecessary" is the operative word. Some EBITDA exposure is unavoidable — macro demand shifts, regulatory changes, technology transitions moving at their own pace. Stellantis could not have prevented EV adoption from slowing. It could not have prevented tariff headwinds. These were real, and were felt across the industry.
What it could have prevented was the scale of the reversal. The difference between a manageable reset and a €25.4B write-down is not the market. It is the point at which risks were quantified, prioritised, mitigated, and predicted. Earlier detection. Staged optionality rather than binary bets. Supplier contracts with exit terms modelled upfront rather than discovered at cancellation.
NAVETRA™ does not claim to eliminate EBITDA risk. It claims to prevent the unnecessary portion — the gap between what the market did and what governance allowed to accumulate undetected. In the Stellantis case, that gap was €24.6 billion.
The question was never whether Stellantis faced hard conditions. It did. The question is whether €25.4B in adjustments was the inevitable cost of those conditions — or the preventable cost of not having a system to quantify, prioritise, mitigate, and predict execution risk before it reached the balance sheet. The evidence strongly suggests the latter.
The Question for Your Organisation
Every organisation running a transformation carries some version of this exposure — commitments sized against single base cases, operational signals that never become financial ones, workforce and technology bets tracking narrative rather than evidence.
The Stellantis case is unusually large and unusually well-documented. But the pattern is not unusual. The question it raises is the same one every leadership team and board should be able to answer: how much of your current EBITDA exposure is quantified, prioritised, mitigated, and predicted — and how much is sitting where the €25.4B sat, accumulating until the accounting forces you to see it?
Execution risk that isn't measured doesn't disappear. It accumulates — until the accounting forces you to see it.