The non-profit International Council for Clean Transportation in Washington compares real-world results posted on a popular German website with official EU figures and finds the gap grew from 8.0% in 2001 to 21% in 2010.
Researchers used records provided by drivers of 28,218 cars, including BMW 5-Series.
PARIS – The push for a new test-drive cycle in Europe, and the rest of the world for that matter, gets a boost from a study on vehicle fuel efficiency in Germany that shows real-world results are moving further away from official mileage figures issued by the European Union.
The working paper by the non-profit International Council for Clean Transportation, based in Washington, compares real-world results posted on a popular German website with official EU figures and finds the gap grew from 8.0% in 2001 to 21% in 2010.
From the environmental lobby’s point of view, this means actual reductions in carbon-dioxide emissions are not falling as fast as some people may think.
European auto makers use the New European Driving Cycle to homologate their vehicles, which calls for stops, starts, accelerations and decelerations in a precise pattern. But the ICCT and others argue the cycle does not reflect actual driving conditions. Accelerations are milder and gears are changed earlier than what is observed on the roads.
“The fixed speeds, gearshift points and accelerations of the NEDC offer possibilities for manufacturers to optimize CO2 and pollutant emissions specifically for the corresponding operating points of the engine in order to achieve lower emission levels during type approval, but not necessarily for real-world driving,” notes the ICCT in its working paper.
“(There are) discrepancies between type approval and real-world fuel consumption and CO2 values.”
The EU figures show that emission levels for Germany have declined from 179 g/km of CO2 in 2001 to 152 g/km in 2010, a 15% improvement. However, the ICCT says calculations from German drivers show levels fell from 192 g/km to 179 g/km in the timeframe.
A United Nations project is working on a new global mileage test that would allow better comparisons across continents, and it is expected to have a driving cycle closer to actual use.
The ICCT believes the World Harmonized Light-Duty Vehicle Test Procedure could be adopted in Europe by 2015, although that probably is optimistic. Given the massive investments auto makers have made to achieve their best scores with the existing rules, there will be opposition to change.
Henri Trintignac,’s vice president-electrification projects, acknowledges the NEDC does not reflect actual driving conditions. Nor does it adequately reflect the benefits that can be achieved with electrified powertrains. But he says it does permit legitimate comparisons between different vehicles.
The move to establish a global test drive and procedure seriously has been under way since 2009, when the European Commission began looking forward to Euro 6 emissions standards. It began as cooperation between Japan, the U.S. and the EU, with a goal for adoption in 2014.
However, according to groups taking part in the initiative, such as the Swedish Test Center and global-supplier, new elements are being added to the rule preparation, such as evaporative emissions, daylight running lights, the effect of air conditioning and battery state of charge.
says there could be high-speed elements for North America and Europe, and low-speed rules for entry-level vehicles in India.
In its review of global emission standards on its website, Delphi says curtly: “Many delays encountered up until now. Earliest expected application in Europe 2017.”
Real-world results in Europe may differ more from the test figures than in the U.S. The U.S. rating is based on tests of the best-selling version of a vehicle. But in Europe, auto makers test the base version even though most are sold with options that add to the mass and detract from the fuel economy.
The ICCT says the U.S. approach also is better because auto makers have to provide all relevant information for each model year of their production vehicles.
In its study, the ICCT acknowledges the information from German consumers on the website spritmonitor.de is not scientific, but it argues that the number of participants involved give the averages credibility.
ICCT researchers used records provided by drivers of 28,218 cars. The models involved represent more than 50% of annual sales in Germany.
They include the Audi A3, 4, 6 and 8;1-, 3-, and 5-Series and X5; Ka and Focus; Mercedes-Benz C-, E-, ML- and SLK-Class; Opel Astra and Corsa; Renault Clio, Megane and Twingo; Aygo; and VW Polo, Golf and Passat.
Researchers say they will expand their working paper if they can collect usable data from other countries as well.