2.1 Air pollution by particulate matter
Reducing air pollution is one of the focus areas for European environmental policy, as stated in the 6th Environmental Action Program of the European Union. There it says that until 2012, a level of air quality that “does not give rise to significant negative impacts on and risks to human health and the environment” should be achieved (DECISION No 1600/2002/EC). Already in the so-called Air Quality Framework Directive from 1996, Particulate Matter (PM) is included as one of the pollutants to be monitored and regulated in the future (Council Directive 96/62/EC). The term Particulate matter refers to particles that can differ in origin, chemical composition, and size and is classified by its diameter. PM10 denominates particles with a diameter of 10 micrometers (hereafter: µm) or less. It can be emitted directly, or is formed in the atmosphere from the oxidation and transformation of gaseous particles. Fuel combustion, such as for heating, driving, or generating electricity, is the human activity that contributes mostly to the formation of particles (European Environmental Agency, 2011). PM enters the body through the respiratory tract and penetrates into the lungs. The smaller the particles, the deeper it penetrates into the respiratory system. While the coarse fraction of PM10 mainly affects airways and lungs, smaller parts can be absorbed by the blood system and penetrate deeply into lung tissue (In-text citationEuropean Environmental Agency, 2010). [/annotax] The exacerbation of asthma symptoms, pulmonary and cardiovascular diseases, and respiratory cancer are to name but a few health problems connected with PM exposure (World Health Organization, 2006).
2.2 EU legislation on air quality and implications for Germany
In 1999, two binding PM10 limit values were set for all member states of the European Union: The yearly average concentration should not exceed 40µg/m³ and the average concentration measured within 24 hours is not allowed to exceed more than 50µg/m³ on more than 35 days per year. Both limit values entered into force on January 1, 2005 (COUNCIL DIRECTIVE 1999/30/EC). Non-compliance with the limit values can be sanctioned by the European Commission. However , cities can apply for extensions to reach the limit values, if they provide convincing evidence that the limit violation was due to unusual circumstances. If this is not accepted, a financial penalty can be imposed, depending on the seriousness and the duration of the violation, the population affected, and the economic power of the country in question (Wolff and Perry, 2010).
To monitor PM10 concentrations, the member states have to set up a network of measuring stations. In Germany, the measuring and monitoring of local PM10 concentrations falls within the responsibility of the 16 federal states. Each state operates its own measuring network; the data is then collected and validated by the Federal Environmental Agency. General criteria of the placement of measuring stations are set up by EU regulation, but the final locations are determined by the state environmental agencies (COUNCIL DIRECTIVE 1999/30/EC). The stations are classified by the environment they are placed in (urban, suburban, peri-urban, rural, rural-regional, or remotely rural) and the type of pollution they are supposed to measure: Traffic pollution, industrial pollution, or background concentrations of PM10. Between 2005 and 2010, 345 measuring stations in Germany recorded a non-compliance with one of the European limit values. The yearly average of maximum 40µg/m³ was exceeded 18 times, while in 327 cases, a PM10 concentration above 50µg/m³ was recorded for more than 35 days per year.
Table 1 provides an overview of the non-compliances occurring between 2005 and 2010. As can be seen, problems occurred especially with the 24 hour limit value, which was violated up to 102 times a year. Most of the non-compliances occurred at stations measuring emissions from road traffic. Depending on the year, up to 53% of these stations recorded more than 35 days a year with concentrations above 50µg/m³. However , the number of non-compliances varies considerably between the years. This variability can to a large extent be attributed to meteorological conditions that influence ambient particle concentrations. Good vertical exchange of air leads to a better dispersion of particles and transportation to higher atmospheric layers and thus lowers concentrations near the ground. The same goes for precipitation, while high atmospheric pressure, low wind speeds, and dry conditions impede the transportation of particles and thus lead to high concentrations near the ground. Also due to wind and weather conditions, transboundary air pollution transporting particles originating outside of Germany can vary between the years (Federal Environmental Agency, 2009).
Table 1: Exceedance of PM10 limit values in Germany, 2005 – 2010 [Table not shown]
2.3 Low emission zones as a policy tool to cut back air pollution
European air quality regulation does not only set limit values for particle concentrations, but also defines procedures that have to be followed when a member state is in non-compliance with a limit value. Each member state has to divide its territory into management zones. If a PM10 limit is exceeded in a zone, the responsible regional administrative entity has to establish an air quality plan (AQP) for that zone no later than two years after the violation occurred. AQPs have to specify measures and strategies on how to prevent future non-compliance, which the cities are free to choose (DIRECTIVE 2008/50/EC). Most often, tools of urban and regional planning are used, of which a low emission zone is one example, but there is no obligation to introduce an LEZ as a part of an AQP (Federal Environmental Agency, 2012a). Thus , not all cities that set up an AQP decide on introducing an LEZ, but all cities that are in non-attainment with one of the limit values have to set up an air quality plan.
The low emission zone is a tool that aims at reducing PM10 emissions specifically from road traffic. Currently, it is applied by 55 of the 129 cities where an AQP is in place (Federal Environmental Agency, 2012e). It usually takes one to three years to implement it after the AQP has come into force, due to the administrative processes that necessarily precede the implementation of an LEZ4. A low emission zone is an area that can only be entered by vehicles meeting certain exhaust standards and is marked by special traffic signs. If a vehicle fulfills the standard it is marked by a sticker that has to be placed in the front window. A vehicle can either get a green, a yellow, a red or no sticker at all. Which sticker a vehicle qualifies for depends on the emissions group it belongs to. The definition of the emissions group is based on the EURO emission standard5 that is recorded in each vehicle’s registration papers. Table 2 provides an overview about the correspondence between EURO emission standards, emissions groups and sticker color:
Table 2: Overview on emission standards and sticker colors [Table not shown]
Hence, the sticker color determines if a vehicle is allowed to enter the low emission zone or not. Currently, three different levels of low emission zones exist: An LEZ step 1 bans only vehicles that do not have a sticker at all, an LEZ step 2 bans vehicles with a red sticker or no sticker, and an LEZ step 3 bans vehicles with a yellow, red, or no sticker. The implementation and monitoring of the zones is within the responsibility of the regional administrative entities. If a vehicle enters an LEZ without the required sticker, a fine of 40€ and a penalty point6 are imposed (Wolff and Perry, 2010). The Federal Environmental Agency (2012a) expects the introduction of an LEZ step 1 to reduce particle concentrations by around 2% and the number of days with concentrations above 50µg/m³ by 5 per year, while the LEZ step 3 is expected to lead to 10–12% lower average concentrations 10 days less in exceedance of the 24h limit.
2.4 Previous experiences with low emission zones
Germany is not the first country trying to combat air pollution from road traffic through implementing some kind of low emission zone. The following section shortly reviews previous empirical findings concerning the effectiveness of LEZs, which vary by exact design. Carslaw and Beevers (2002) classify LEZ designs into approaches that restrict vehicle use based on current air quality, based on a technology standard, or based on a transport criteria.
A low emission zone using the transportation criteria was for example established in Mexico City in 1989. Based on the last number of the license plate, vehicles were prohibited to enter the Mexico City metropolitan area on a specific weekday. This ban did not lead to any improvements in air quality, but had the reverse effect of increasing pollution levels during weekends and at times where the ban did not apply. It was found that instead of driving less, many people bought a second and often older car to obtain a second license plate, thus not reducing their travels, but instead increasing emissions by using a dirtier car (Davis, 2008). An example for an LEZ based on a technology standard is the so called “Environmental Zone” that was introduced in Stockholm in 1996. Buses and trucks that do not fulfill a certain emission standard are prohibited from entering the city center, while there are no restrictions to passenger cars. Rapaport (2002) finds a 25% reduction in particle emissions following the introduction, which in turn decreases concentration levels by 4%. A similar approach was taken in London in 2008, where an LEZ bans some heavy-duty vehicles, buses, large vans and minibuses from the city centre based on an emission standard. Jones et al. (2012) conclude that following the announcement and introduction of the LEZ, particle concentrations decreased by 30 – 50% depending on the site. However , they admit that this effect might to a large extent be caused by a simultaneously introduced fuel standard.The impact of German LEZs on air pollution has been studied for some of the bigger cities where the policy was introduced at an early stage. In general, two types of studies exist. The first type uses modeling techniques to assess the changes in particle emissions after the introduction of an LEZ. Such a study has been done recently for an LEZ step 3 in Berlin, which was introduced in January 2010. It uses observed changes in the composition of the vehicle fleet and specific emission factors for the sticker colors to calculate the change in PM10 emissions in comparison with a reference scenario. It concludes that the LEZ step 3 reduced yearly average PM10 concentrations by approximately 2µg/m³ at typical Berlin traffic spots compared to an LEZ step 1 and that ca. 10 days of violation of the 24h limit value were avoided (Federal Environmental Agency, 2011). The second type of studies uses observed PM10 concentrations to evaluate the effect of a low emission zone by before-after comparisons. Such a study was done in 2009 for the city of Munich, which introduced an LEZ step 1 in October 2008. Comparing concentrations between October 2007 and January 2008 with the values in the same period one year later, it finds a reduction in concentrations between 5.4 and 12.3% for the measuring points within the LEZ (Cyrys et al., 2009). The same methodology was applied by the city of Bremen in 2009. This study found increases in particle concentrations of 5.2% and 1.9% at two different measuring spots. The city of Hannover found a reduction of average PM10 concentrations by 4% or 1µg/m³ in comparison to the previous year in a report on the LEZ step 1 that was introduced in 2008 (Ministry for the Environment, 2009).
However , both study designs exhibit shortcomings with regard to the causality from policy implementation to PM10 concentrations. The modeling approach might miss out important factors, such as unexpected behavioral changes, and relies on assumptions when constructing the baseline scenario that might not be fulfilled. Before-after comparisons, on the other hand, do not take into account underlying trends in PM10 concentrations. In attributing changes completely to the LEZ, it is assumed that the concentrations would have remained stable in absence of the intervention. However , technological or behavioral change might cause an underlying trend in PM10 concentrations independent of the policy implementation. Thus , for a causal evaluation of LEZs, an empirical strategy that is able to handle both underlying trends and potential behavioral changes is needed. Before presenting the empirical setup used in this study, the following section will provide a theoretical analysis of LEZs as a policy tool.
4 For example, the city of Heidelberg introduced an LEZ in 2010 and considered a period of at least 12 months between the passage and the implementation of the LEZ as necessary. The design and production of traffic signs and the possibility for citizens to adapt to the new regulation are some of the reasons stated for this time lag. See: AQP city of Heidelberg (2006), p. 12 (http://www.rp.baden-wuerttemberg.de/servlet/PB/menu/1187487/index.html).
5 The Euro standards for vehicles are defined upon end-of-pipe exhaust values for a number of exhaust products, such as carbon monoxides, nitrogen oxides, or particulate matter.
6 Eighteen penalty points can lead to a loss of the driver’s license in Germany.