PUBLIC POLICY · TAXATION & ACCEPTABILITY
Europe's climate targets require a sustained rise in carbon taxation on fuel. France must prepare a multi-year trajectory for gasoline and diesel, with an estimated impact of 6 to 8% on pump prices over three years and an expected yield of more than €4 billion a year once fully phased in.
For the state, it's a price signal. For part of the working population, it's the cost of a commute they cannot necessarily do without.
THE PROBLEM
It also says who has to make the effort. The technical decision is sound, the climate ambition is set at European level, and the funding is in place: part of the revenue is earmarked for a transformative mobility plan — rural public transport, vehicle-conversion aid, and expanded regional rail.
It runs into a precedent: the major social crisis of late 2018, triggered by a comparable rise in carbon taxation, paralyzed the country for several months. Since then, every fiscal announcement on fuel has been scrutinized by political analysts, economic journalists, organized social movements, and parliamentary opposition.
The minister's chief of staff mobilized our system to test every combination of compensation schemes under consideration, with a 12-week projection of media and political cascades and fine-grained territorial segmentation. Result expected within four weeks, ahead of presenting a structured file to the Prime Minister.
A DISTINCTION: AN AMOUNT IS NOT A POSITION
Raising the cost of carbon changes behavior and speeds up the transition. But the political reaction can turn on a different question: why me. The problem then becomes a combination of price, ability to adapt, territory, perceived fairness, and political trust.
A MECHANISM: THE OBJECTIVE COST DOESN'T EXPLAIN EVERYTHING
The working adults who reject the measure most fiercely are not those who would pay the most. 68% of the intensity of the rejection is explained by a sense of political exclusion, independent of the actual tax amount borne.
The analytical point isn't a moral one: a fiscal measure can be read as a signal of one's position in society. A compensation calculated on the burden alone doesn't address that part of the rejection.
A MECHANISM: ABILITY TO CHANGE IS NOT AN INCENTIVE TO CHANGE
The same economic signal changes nature depending on what's accessible: available public transport, ability to replace the vehicle, commuting distance. Where an alternative exists, the increase is an incentive. Where it doesn't, it's an added cost with no possible change.
This is what makes earmarking revenue toward the alternatives themselves — rural public transport, conversion aid, regional rail — inseparable from the fiscal scheme.
A MECHANISM: TERRITORY IS A VARIABLE OF THE POLICY
The 24 territories were defined by cross-referencing density, median income level, employment structure, vehicle-ownership rate, and political exposure to mobilizations over the past five years.
A single national rule runs through these 24 contexts and produces different consequences in each. An average national population would have masked the five territories that carry most of the risk.
WHAT WAS TESTED
The question wasn't “tax or no tax.” It was: what architecture of tax and compensation can hold, in which territories, in what order.
The families of schemes documented in this case. Each of the 9,600 synthetic citizens interviewed was exposed to the 56 schemes in variable combinations.
The reconstructed population covers 8.2 million French working adults who use personal vehicles, calibrated on public data from INSEE, the Ministry for the Ecological Transition, the Commissariat général au développement durable (the French general commission for sustainable development), and national mobility surveys. No nominative data was entered into the system.
It is structured across 24 distinct territories, defined by cross-referencing density, median income level, employment structure, vehicle-ownership rate, and political exposure to social mobilizations over the past five years. Territorial consistency was validated by cross-reference with INSEE's territorial studies and the public mappings from the Directorate General for Local Authorities (DGCL).
DENSITY AND EMPLOYMENT STRUCTURE OF THE TERRITORY
two of the five variables used to define the 24 distinct territories.
MEDIAN INCOME LEVEL
what determines the share of the fuel bill within the constrained budget.
VEHICLE-OWNERSHIP RATE
from complete car dependency to occasional use of a personal vehicle.
POLITICAL EXPOSURE TO MOBILIZATIONS OVER THE PAST FIVE YEARS
what differentiates the speed at which discontent becomes a movement.
These configurations make part of the population's heterogeneity visible. The simulation is based on synthetic individuals, not on a handful of persona types.
REACTIONS
THE REJECTION IS NOT FISCAL, IT IS SYMBOLIC
The working adults who reject carbon taxation most fiercely are not those who would pay the most. They are those who don't feel part of the transition. 68% of the intensity of the rejection is explained by this sense of political exclusion, independent of the actual tax amount borne. Financial compensation isn't enough: the sense of exclusion from the decision-making process must be addressed, which requires a participatory scheme, not just a redistributive one.
THE NATIONAL FUEL VOUCHER BACKFIRES IN RURAL AREAS
A uniform national fuel voucher achieves 62% acceptance in urban areas and 32% in rural areas. Same beneficiaries, same amount. In rural areas it is read as a symbolic compensation meant to make an unfair-feeling tax acceptable. The same amount, structured as a territorial mobility grant and rolled out at the departmental level with local elected officials, achieves 71% acceptance in the same areas.
THE ORDER OF ANNOUNCEMENT DETERMINES 80% OF ACCEPTANCE
Six sequencings were tested. Announcing the tax before the compensations triggers a media crisis in 87% of the simulated scenarios. Announcing the compensations before the tax, even three weeks ahead, flips the reading: the tax becomes the funding of a mobility plan, not a punishment with a bandage attached. The order of announcement weighs more than technical adjustments to the amounts.
THE 24 TERRITORIES DO NOT REACT AT THE SAME SPEED
In eight metropolitan territories with a strong public-transport network, acceptance stabilizes within three weeks. In eleven peri-urban or rural territories, the trajectory spreads over eight to twelve weeks. In five low-density rural territories with a history of mobilization, it only stabilizes after 15 weeks and remains 20 points below the national average, even in the best-case scenario.
TRANSPARENCY ON FUND ALLOCATION IS WORTH MORE THAN A LOWER TAX
Two strategies with equivalent yield were compared: a tax at 6% of the pump price with no allocation commitment, and a tax at 8% with a binding commitment to allocate 100% of revenue to the mobility plan. The second, though costlier for the driver, achieves +18 points of acceptability. The visible, controllable counterpart changes the nature of the measure.
PREFECTS ARE THE CHANNEL OF TERRITORIAL CREDIBILITY
A national announcement via the government spokesperson achieves an initial acceptability of 34% on average across territories. A cascading announcement, with each prefect communicating the scheme in their department together with local elected officials in the week before the national announcement, achieves 61%. The scheme is then received as a negotiated agreement for the territory, not as a decision imposed from Paris.
A DISTINCTION: DISTRIBUTIVE JUSTICE, PROCEDURAL FAIRNESS
The administrative intuition is simple: tax, plus an equivalent voucher, equals problem solved. Yet the uniform national fuel voucher achieves only 32% acceptance in rural areas, against 62% in urban areas, for identical amounts and beneficiaries.
The first question is “how much do I pay, how much do I get back.” The second is “how and why was this decision built.” The same amount, structured as a territorial mobility grant and rolled out at the departmental level with local elected officials, achieves 71% acceptance in the same rural areas.
The narrative structure of the scheme therefore counts as much as its financial content. It is the same sum of money, received within two different relationships to the decision.
A TERRITORIAL MECHANISM: SAME INCREASE, DIFFERENT EFFORT
The media rollout of the announcements produces very different trajectories across territories. This isn't a binary opposition between city and countryside: it's a distribution of contexts, five of which concentrate most of the mobilization risk.
A TEMPORAL MECHANISM: OPINION IS NOT STATIC
The likely media and political cascades were simulated over 12 weeks for each scenario, modeling the relays of French political debate: national daily press, rolling news channels, social media, unions, parliamentary opposition, road-user associations, organized protest movements.
Individuals observe what others think and how those in power respond. Initial media visibility can legitimize further protest, and the acceptability trajectory by territory then reveals irreversible tipping points.
A DISTINCTION: MATERIAL VULNERABILITY, RISK OF PROTEST
The populations that are most economically constrained, most politically distant, and most capable of mobilizing do not overlap. This is why political exposure to mobilizations over the past five years was chosen as a variable for defining the territories, on a par with median income and vehicle-ownership rate.
Five low-density rural territories with a history of mobilization concentrate the risk: their acceptance only stabilizes after 15 weeks and remains 20 points below the national average, even in the best-case scenario. Prioritizing them, with dedicated local dialogue schemes, secures the whole rollout.
COMPARISON
Only the dimensions for which the case documents measures carry numeric values. The “not disclosed” label is kept rather than an estimate.
TAX AT 8% WITH BINDING ALLOCATION
Visible, controllable counterpart → +18 pts of acceptability
TAX AT 6% WITH NO ALLOCATION COMMITMENT
Lower levy → lower acceptability for equivalent yield
A DISTINCTION: ACCEPTABILITY IS NOT A PROPERTY OF THE MEASURE
The same compensation scheme achieves 32% or 71% acceptance depending on how it is structured and by whom it is carried. The same announcement achieves 34% or 61% initial acceptability depending on whether it comes through the government spokesperson or through prefects and local elected officials.
Acceptability therefore depends on the measure, but also on the population, the territory, the alternatives available, the compensation, trust, and the sequence of the announcement. It can be designed, not merely measured.
This case adds a complete chain to the library: measure, populations, territories, interpretations, adaptations, mobilizations, media cascades, political reaction, new trajectory.
THE ACTUAL ROLLOUT
The inter-ministerial committee adopted the dominant scenario: compensations announced three weeks before the fiscal announcement, a territorial mobility grant rolled out at the departmental level, a binding commitment to allocate 100% of revenue to the transformative mobility plan, and a media cascade through prefects in the week before the national announcement.
The plan was approved by the Prime Minister after six weeks of inter-ministerial negotiation, with the fiscal increase ultimately raised to 8% instead of the 6% initially envisaged, to fund a more ambitious mobility plan than originally planned.
Rollout began five months after the decision, once agreements with local authorities had been negotiated and compensation schemes calibrated by department.
ARBITRATION
PROJECTION, THEN OBSERVATION
Over the first 12 weeks, no major political crisis was triggered. Two limited territorial mobilizations took place, in two of the five most historically mobilized rural territories, and were resolved through a targeted acceleration of local dialogue schemes. National press coverage remained largely descriptive, with criticism focused on technical parameters — the grant scale, eligibility thresholds — without a frontal challenge to the principle.
At 18 months after rollout, national acceptability of the carbon tax rose from 41% before the announcement to 67% after full deployment of the compensations. In the five initially high-risk rural territories, it rose from 22% to 54%, beyond the initial projection of 48%.
Fiscal yield is in line with projections. The transformative mobility plan is being rolled out in 18 of 22 regions, broadly on schedule. The executive preserved its capacity to speak on climate policy without having to absorb the cost of a major social crisis.
These figures describe what was observed after the decision. They establish no exclusive causality between the architecture adopted and any of these movements.
LESSON
It was who felt they were paying for a decision made without them. 68% of the intensity of the rejection is explained by a sense of political exclusion, independent of the amount borne: a public policy that distributes costs and benefits implicitly says who must make the effort.
This is why the most powerful levers identified in this case are not about the amount: the order of announcement, the level of government carrying the scheme, the visible allocation of revenue. The same compensated euro does not have the same effect depending on the relationship it establishes between the decision and those who bear it.
POSSIBLE FUTURES
A
UNIFORM PRICE SIGNAL
An identical increase everywhere, with no associated compensation architecture
B
TAX AND MONETARY COMPENSATION
Uniform national fuel voucher, same amount for all beneficiaries
C
TERRITORIALIZED TRANSITION
Compensations announced before the tax, departmental territorial grant, full allocation of revenue, prefectoral cascade
METHOD
8.2 million French working adults who use personal vehicles, calibrated on public data from INSEE, the Ministry for the Ecological Transition, the Commissariat général au développement durable (the French general commission for sustainable development), and national mobility surveys, structured across 24 distinct territories. 9,600 synthetic citizens were interviewed individually, distributed proportionally across the 24 territories, each exposed to the 56 compensation schemes in variable combinations. Dynamic agents followed up with each citizen on friction points identified in real time, with an average of 9 follow-up questions per person.
The likely media and political cascades were then simulated over 12 weeks for each scenario, modeling the specific relays of French political debate. This modeling produced an acceptability trajectory by territory and by scenario, with identification of irreversible tipping points. It is this combination — territorialized individual listening and simulation of political cascades — that made it possible to identify the architecture of announcing compensations first, territorial grant, prefectoral dialogue as dominant in 21 of 24 territories.
This case comes from a simulation carried out for a sovereign ministry facing a politically sensitive execution decision. The client is not named, no nominative data was entered into the system, and detailed results remain the client's property.
The simulation does not substitute for democratic decision-making, does not say which policy should be adopted, and is not used for individual political targeting. It exposes the possible consequences of several taxation and compensation architectures across different populations and territories. The normative arbitration remains the decision-maker's.
Your next decision
Describe your need. We can point you to the right level of support.
State your decision. See the future it produces.
