Congestion pricing, air pollution, and individual-level behavioural responses

Combatting climate change and poor urban air quality will require a fundamental shift towards greener modes of transport. Policies that incentivise individuals to choose low-emission alternatives to conventional transport will likely play a key role in this transition. This paper examines the example of congestion charging and shows that differentiating driving costs by time of day and vehicle type can help to improve urban air quality, reduce driving and induce adoption of electric vehicles.

The authors find that a congestion charge imposed in Bergen, Norway, that imposed spatial and temporal variation in the cost of driving a conventional vehicle, led to reduced volumes of traffic and concentrations of nitrogen dioxide (NO₂). Households exposed to congestion charging on their way to work were more likely to adopt an electric vehicle (EV). Overall, the findings on car ownership suggest that congestion charging combined with exemptions for EVs can be a powerful tool to promote EV adoption, but that there are systematic differences in how households respond to the policy, depending on factors including income, level of education and access to public transport.

Key points for decision-makers

• The authors combine highly detailed data on air pollution, traffic and car ownership to shed light on the efficiency and equity impacts of a congestion charge that increases the costs of driving petrol/gasoline and diesel vehicles during rush hours.
• The data relate to a congestion charge implemented in 2016 in Norway’s second largest city, Bergen, which at the time of the study raised the price of entering the city centre toll cordon during rush hours by 80%. The charge only applied to weekdays, and only to petrol and diesel vehicles.
• Increasing the charge of entering the toll cordon during rush hours (6:30–9am and 2:30–4:30pm) by around 80% led to a 14% decrease in cars entering the congestion zone at these times and an 11% reduction in concentrations of NO₂ around midday.
• The findings suggest that drivers primarily substituted towards other modes of transport rather than simply changing the time they were driving.
• The results also suggest that households respond to the congestion charge by replacing their petrol or diesel vehicles with electric vehicles (EVs). Households exposed to the Bergen congestion charge were around 4.2 percentage points more likely to purchase an EV than those in Stavanger, a city without a congestion charge chosen for comparison.
• The authors also find that patterns of EV adoption diverge along several socioeconomic dimensions. While the policy had no effect on EV adoption among households in the lowest income quintile, which may be due to the comparatively high cost, the EV share for households in the highest income quintile increased by around 7 percentage points as a consequence of the policy.
• University-educated couples with children, and households with a longer work commute and whose public transport options were poor, were also more likely to buy an EV, with the latter implying that the quality of alternative transport options plays a key role in how households adapt to the charge. 
• The results should be considered in the Norwegian context, where financial incentives have contributed to an exceptionally high market share of EVs and a relatively well-developed charging infrastructure. However, the findings may help shed light on expected impacts of congestion charging in other countries in a future scenario where EVs are more competitive with internal combustion engine vehicles.