Introduction

Governments frequently implement policies intended to improve equity that inadvertently weaken environmental incentives. One striking example arises in the transportation sector, where many nations adopt vehicle purchase taxes or registration fees that decline with vehicle age, making older cars relatively cheaper to operate. This research investigates this trade-off in Brazil, where age-based exemptions from the vehicle ownership tax (IPVA) create incentives to retain older, more polluting vehicles. By exploiting variation in exemption thresholds across states, the study examines their effects on vehicle fleets, pollution, health, fiscal outcomes, and political behaviour.

Institutional background

Brazil’s vehicle ownership tax is an annual ad valorem tax on motor vehicles. Revenue is split equally between the state of registration and the municipality of the owner’s residence. During the 2013-2020 study period, 25 of 27 states exempted vehicles once they reached a predetermined age threshold. While these exemptions are typically motivated by equity considerations — intended to promote ‘fiscal justice’ and eliminate distortions penalising the poorest families — they reduce the relative cost of keeping older vehicles on the road. This design contrasts with the logic of an optimal pollution-mitigation policy, such as a Pigouvian tax, which would instead impose higher costs on more polluting vehicles to reflect their environmental externalities.

Conceptual framework

The report develops a framework to illustrate how tax policy affects households’ dynamic replacement decisions. A discrete drop in taxes at an age threshold creates a dynamic incentive to delay vehicle replacement. Forward-looking households form expectations of future tax relief, leading to ‘bunching’ in the vehicle age distribution just above the exemption threshold. This shift towards an older fleet leads to higher aggregate emissions, as older vehicles generally have higher emission rates due to less stringent historical standards and the natural deterioration of emission control systems with age and usage.

Data

The analysis utilises public municipality-level vehicle registration data from the National Registry of Motor Vehicles (RENAVAM) covering the period from 2013 to 2020. Pollution levels are measured using municipality-year estimates from the Greenhouse Gas Emissions and Removals Estimating System (SEEG) and satellite-based measurements of PM2.5 and carbon monoxide concentrations from MERRA-2. Administrative health data on infant birth and death records are sourced from the Brazilian Unified Health System Database (DATASUS). Electoral data for gubernatorial and state assembly elections are obtained from the Superior Electoral Court.

Border-pair approach

To identify the causal effects of the tax policies, the research employs a border-pair design. This approach compares neighbouring municipalities on opposite sides of state borders that are subject to different exemption policies. By restricting the analysis to geographically close groups, the study holds constant many local characteristics, assuming that assignment to either side of the border is as good as random. The analysis categorises exemption policies into three groups based on vehicle age thresholds: states with a cutoff of 10 years, states with cutoffs between 15 and 20 years, and states with de facto no-exemption policies (thresholds at 28 years or higher).

Main results

The research reveals that exempting vehicles above thresholds of 10 to 20 years leads to a significantly older fleet. Municipalities in states with a 10-year cutoff have average vehicle ages approximately one year higher than the baseline group. This anti-scrappage effect increases pollution; municipalities with a 10-year cutoff show on-road CO2 emissions per capita approximately 23% higher and PM2.5 concentrations 4% higher than the baseline. These environmental impacts correlate with worse infant health outcomes, including lower average 5-minute Apgar scores (reduced by 0.17 in the 10-year group) and higher rates of low birth weight (increasing by 8.2 per thousand live births). Heterogeneity analysis suggests the policy’s equity benefits are negligible, as no evidence was found of higher vehicle ownership per capita among lower-income families.

Political economy and fiscal implications

The widespread adoption of low exemption-age policies appears to reflect local political incentives. Using state reforms in Goiás and Mato Grosso do Sul that raised exemption ages, the study finds that politicians associated with such changes are electorally penalised. For each percentage point of the fleet share that lost exemption status, gubernatorial vote shares for the incumbent party declined by 2.3 percentage points. Raising exemption age thresholds by five years increased annual consolidated IPVA revenue by $35.8 BRL per vehicle. Combining these margins reveals a sharp political-fiscal trade-off: each 1% increase in revenue is associated with a 1.5-percentage-point loss in vote share. This dynamic creates a ‘political environmental trap’ where environmentally harmful policies become electorally difficult to reverse.

Concluding remarks

The Marginal Value of Public Funds (MVPF) for lowering the exemption age threshold is estimated at approximately -0.78, meaning the policy generates a net welfare loss of $0.78 for every $1 of forgone government revenue. The findings suggest that vehicle taxation should account for both environmental costs and distributional consequences. Rather than subsidising the continued use of older, more polluting vehicles, governments should direct fiscal support towards the purchase of newer, cleaner models. Gradual reduction of age-based exemptions, combined with targeted support for lower-income households and taxes linked to environmental performance — similar to the MOVER and Move Brasil programmes — would better align taxation with sustainability goals.