In the wake of the 2015 Volkswagen testing scandal, research by TSE’s Mathias Reynaert suggests that emission standards can be a risky and unpredictable policy tool. Investigating strategic responses in the European car market, he finds a growing divergence between on-road fuel consumption and laboratory results that suggests widespread ‘gaming’ of tests by manufacturers.
Today all major vehicle markets have adopted emission standards to improve local air quality and/or to regulate the production of greenhouse gases. In 2007, the EU announced one of the world’s most demanding policies, obliging automakers to cut air pollutants by 18%.
Evaluating the welfare impact of emission standards is not easy. It requires consideration of the political environment, the enforcement of the policy, and strategic decisions by firms. In a new paper ‘Abatement Strategies and the Cost of Environmental Regulation’, Mathias discusses the following responses that firms may adopt:
- Pricing: Firms can change pricing to shift the sales mix to vehicles with CO2 emissions below the target.
- Downsizing: Firms can sell smaller and less powerful vehicles that are more fuel efficient.
- Innovation: Firms can improve the fuel efficiency of their vehicle fleet by adopting technologies that improve the combustion process.
- Gaming: Firms may reduce emissions during the regulator’s tests but not necessarily on the road. Enforcement of the emission standard plays a role in limiting gaming.
Using a detailed panel of vehicle attributes, prices, and sales for the EU market, Mathias finds no evidence of price changes or downsizing in response to the emission standard. Every year, automakers seem to make vehicles that are more powerful, accelerate faster, and are larger, while emissions do not increase. The same pattern of technological progress has been observed in the US market. If automakers use these advances to make more fuel-efficient vehicles, firms should be able to comply with emission standards. In the EU market, technological improvements appear to have happened twice as fast after the announcement of the emission standard. However, this is the result of looking at official emission numbers obtained from laboratory tests.
In a forthcoming paper, Mathias and his co-author compare the laboratory ratings, which form the basis of policy, with direct measures of on-road fuel consumption. They construct a data set that tracks fuel consumption and kilometers travelled for a panel of more than 250,000 drivers for 12 years in the Netherlands. Using these data, they estimate the percentage difference between the laboratory test and on-road performance for each vehicle vintage and model.
The effects of gaming
What are the welfare effects of emission standards when compliance strategies are technology adoption and gaming rather than price changes? And why did the market respond in this way to the EU standard?
Because of technology adoption, firms' costs increase. The increase in costs reduces profits and consumer surplus. Because of gaming, the reductions in actual CO2 emissions are just 5% instead of the 18% target. The combined value of emission savings and consumer and profit losses is negative. However, when Mathias considers two additional non-targeted welfare effects, he finds the emission standard to have a small positive impact as it also reduces other externalities, such as local pollution, congestion, and accident risk.
What if the EU designed the regulation differently? Using his model to analyze alternative market outcomes, Mathias focuses on the standard’s attribute base and lack of enforcement.
Attribute basing makes the emission target dependent on vehicle weight. Firms selling lighter vehicles face a more stringent target. He finds that attribute basing makes it much costlier to lower emission by changing prices. Firms have to distort prices more to reach the target because there are fewer vehicles to which firms can shift sales. If the regulation has a flat target without attribute basing, firms opt for changing prices together with some technology adoption. The flat target reaches actual CO2 emission reductions of 11%, much closer to the 18% target.
The introduction of attribute basing redistributes the incidence of the regulation between French, Italian, and German producers. Mathias’ simulations show that the positions of the national governments are in line with the interests of their domestic firms. The French and Italian governments were in favor of regulation without attribute basing, while Germany lobbied for a steep attribute design.
Gaming is also a product of the political environment. A recent evaluation by the European Parliament blamed enforcement failures on car-producing member states. A better test procedure would mean that official and actual emissions are more similar. With more enforcement, firms have to adopt costlier technology, and this increases consumer prices. But enforcement would have led to much higher CO2 and other externality savings, and the policy would have been welfare improving.
This research demonstrates that emission standards can be an unwieldy policy tool. The European political environment led to failures in both the design and enforcement of the emission standard that caused startling increases in strategic gaming.
TSE Mag #21 Hiver 2021