The Cato Review of Business & Government
The Cato Review of Business & Government
Ramiro Tovar Landa is an Economic Advisor to the
Executive Secretary Mexican Federal Commission on Competition.
Due to urban development and growing industrialization, air pollution has become a major concern in Latin American cities such as Mexico City, Sao Paulo, and Santiago. Eighty-one million Latin American city-dwellers, 26 percent of the total urban population, live in cities with major air pollution problems. The industrial sector, often concentrated in the cities, and vehicle emissions are the main sources of air pollution.
Urban pollution problems affect the health and the productivity of the 20 million people of the Metropolitan Area of Mexico City (MAMC). Population increases and economic growth in Mexico will put increased pressures on the environment, especially in urban areas, in the years to come.
For those reasons, urban environmental concerns in Mexico require innovative policy responses. In fact, the Salinas administration made environmental policy one of its highest priorities in the 1989-1994 National Development Plan. Mexico's recent environmental policy actions have largely followed the American model, implying extensive reliance on command-and-control regulation. The Mexican government's response may be understandable, given the emergency nature of the programs recently announced, but it is not clear that it is the most efficient policy response.
Although relatively unexplored by regulators, applying the concept of tradable permits to mobile pollution sources could improve the air quality in Mexico City in a more cost-effective manner. But before one can examine the case for such a permit system, one must be familiar with the current air pollution situation in the MAMC, the current policies toward air pollution, and their outcomes. Only then can the alternative policy options be discussed.
Air Pollution and Policy in Mexico City
Air pollution has been linked to morbidity and premature mortality, crop damage, forest damage, surface water acidification, and materials and building corrosion. High atmospheric lead levels result in elevated lead concentrations in blood and have been associated with decreased cognitive function in children and increased blood pressure in adult males.
The air pollution problem in the MAMC is very severe, sometimes characterized as the worst in the world. Geographic factors exacerbate the problem. For instance, the MAMC is situated at a high altitude, which causes incomplete fuel combustion. Also, the MAMC is inside a valley, so the air circulation is impeded by mountain ranges.
In the MAMC, as in most metropolitan areas in the world, much of the emissions of ozone precursors is vehicular in origin. The MAMC has a fleet of approximately 2.9 million vehicles, with an average age of nine years; only the models built since 1991 have catalytic converters.
The contributions from mobile emissions sources are 60 percent of the total emissions, with 50 percent of all mobile sources from private autos, 25 percent from public transportation, and the remainder from gasoline and diesel-powered trucks. The contributions by the transport sector to the total emissions level are: nitrogen oxides (NOx), 55 percent: hydrocarbons (HC), 83 percent; sulphur dioxide (SO2), 12 percent; particulates (TSP), 12 percent; carbon monoxide (CO), 98 percent; and lead, 100 percent.
Because of population pressure, the public transportation system, in spite of its recent expansion, lags behind the growing transportation demand. That increases the incentives for automobile ownership; in fact, of the 29.5 million trips made daily, 15 percent are made in private autos, which are 95.4 percent of the total vehicles in circulation. Auto ownership adds to traffic flow congestion, lowering the average speed and increasing pollution levels.
To figure out the magnitude of the effect air pollution has on public health, it is necessary to know the difference between actual air quality and air standards, since the latter are typically set below some threshold related to health damage. Records show that ozone originating from mobile sources, the harmful health effects of which are well known, regularly exceeds the Mexican one-hour ozone standard of 0.11 ppm-sometimes by as much as 400 percent. The eight- hour standard for CO is 0.13 ppm, and the highest concentration measured in the MAMC is 24 ppm. The Mexican NOx one-hour standard is 0.21 ppm: the highest recorded concentration is 0.32. The U.S. lead standard of 1.5 ug/m3 for a three-month period was exceeded in four out of eight quarters in 1986-87. The 24-hour SO2 average is 0.13 ppm and has been declining as the result of fuel switching in electric power generation and industrial use. Concentrations of other toxic air pollutants, such as benzene, diesel particles, and formaldehyde, all produced by motor vehicles, are probably also high, but still are not monitored adequately. The departures from standard levels are particularly severe in the winter months, when temperature inversions and calm winds occur regularly; however, the worst month is May, when solar radiation is at its highest level.
Mexico City is a clear example of a city with multivariable pollution phenomena. Unlike other pollutants, smog is formed through chemical reactions between emissions of volatile organic compounds and nitrogen oxides. That process requires sunlight, high temperatures, sufficient concentrations of pollution, and appropriate wind patterns-highly variable factors. Pollution from mobile sources in particular is a function of weather and tailpipe emissions, an important matter to consider when examining policies for pollution abatement.
Current Policies
The policies adopted to address the air quality problems in Mexico City have been command-and-control strategies involving specified emissions technologies for vehicles and quantitative restrictions on vehicle use. A sophisticated automatic monitoring network has been established to provide information about the air quality each hour in the five different areas that comprise the MAMC. Emissions testing and engine tune-ups are required of all passenger vehicles, though the requirements can often be evaded through bribery. Improved diesel engines have been installed in the bus fleet, and over the next three years the public transport fleet will be converted to use butane gas. In addition, the lead content of gasoline is being reduced to European Community standards, and low-sulfur fuel oil is being forced on industry. Perhaps most important, the government has designated no-driving days for all vehicles. Other measures that have been taken include the closure of a refinery inside the MAMC, a mandatory 50 percent reduction in cab use on Saturdays, and a requirement that firms switch to natural gas or reduce their operations by 30 percent in 1994.
In March 1994 a new program ordered 220 of the top industrial firms, the source of 8.4 percent of the air pollution, to cut their particulate and volatile organic pollutants by 90 percent in 18 months and reduce emissions of reactive organic gases by 40 percent within one year, or quit the city by the second half of 1994. The government is offering 20-year credits to finance either the cleanup or relocation of affected facilities.
In addition, there are emergency plans-referred to as "Phase I" and "Phase II"-used in the pollution alert periods that are declared when the air pollution index rises over 300 points. The emergency plan in Phase I grounds half of the municipal fleet, suspends activities such as paving, and requires that major stationary sources reduce their activities by 30 percent. Phase II requires an additional no-driving day plus further restrictions of industrial activity and use of official vehicles.
The No Driving Day Program (NDD) involves not driving one day during the week (weekends excepted) and two days during Phase II (weekends included). The program was implemented in November 1989 for the winter season.
The immediate results from the NDD were an effective reduction of 20 percent of the vehicles in circulation, an increase of 8 kph in traffic speed, a five-minute reduction in mean trip time, a decline in the rate of gasoline consumption (from 7.6 to 1.2 percent), and an increase of 6.6 percent in subway ridership. Mexico City authorities advertised the result and various environmental groups applauded it, so the authorities decided to make the program permanent.
However, once the program was made permanent, it led to substantially different consumer behavior than its advocates predicted. The driving public, faced with a long-run restriction on driving, apparently found the public transportation system a very imperfect substitute. So instead of flocking to public transportation as the authorities had hoped, residents of Mexico City simply purchased more vehicles, in order to always have at least one vehicle available on any given day. Demand for automobiles shows a high income-elasticity. Estimates indicate that a 1 percent increase in GOP leads to more than a 3 percent increase in automobile demand. The real price of automobiles in Mexico has dropped significantly since trade liberalization. Automobile sales in Mexico are at record levels, with the MAMC share of national vehicle sales at around 53 percent. Auto sales throughout Mexico have increased by 25 percent since 1989.
The economic cost of the NDD program is largely due to the fact that the program makes no distinction among the relative values of various days. Work, leisure, and commercial/industrial delivery travels are treated as if they are of equal value. But of course they are not. The fact that 41 percent of vehicles cited for violating the NDD restriction in the program's first year were commercial delivery vehicles shows that some uses are more highly valued than others.
Indirect evidence suggests that the NDD program may have exacerbated the air pollution problem by increasing vehicle purchases. The moving average of peak ozone measurements has actually increased since adoption of the NDD program-as have the ozone readings, which exceeded 200 points. Moreover, the pollution index grew from 86 days in 1990 to 192 in 1991; and traffic congestion and CO2 have increased during the weekends, forcing the authorities to place more mandatory episodic emissions restrictions on industry and commerce, which has in turn increased the economic cost of the whole pollution control program.
Alternative Approaches to Air Pollution Control
Air quality has improved significantly in most U.S. cities over the past two decades. Nevertheless, air quality in many metropolitan areas still falls short of the health-based standards set by the U.S. Environmental Protection Agency. The U.S. Congress is enacting several additional measures aimed mostly at the volatile organic compounds (VOC) that help to create smog. The measures include possible additional controls on petroleum refineries, chemical plants, and other large industrial facilities, and, for the first time, controls on dry-cleaning establishments, auto paint shops, bakeries, and the like.
In addition, the Clear Air Act Amendments of 1990 will mandate a new round of emissions reductions for all cars and light-duty trucks, and require enhanced vehicle inspection and maintenance programs, as well as the installation of vapor recovery equipment on gasoline pumps, reductions in tailpipe emissions, and the implementation of a program for the use of "clean fuels" such as reformulated gasoline, methanol, or ethanol. When fully implemented by 2005, those controls would add around $19 billion to $22 billion annually to the pollution control expenditures already in place. The economic benefits that would accompany the control measures-such as health and agricultural benefits from reductions in ozone and VOC concentrations-would be in the range of $4 billion to $12 billion per year.
Clearly, the cost of U.S. ozone abatement regulations exceeds the economic and public health benefits by a wide margin. That should not surprise, given that empirical studies comparing the costs of command-and-control programs with those of a more market-oriented regulatory structure indicate that the regulatory cost of command-and-control strategies is often several times that of more flexible regulatory approaches.
Market-based approaches employ economic instruments to take advantage of the large differentials in abatement costs among polluters-the main information problem confronted by regulators under traditional command-and-control approaches. Also, economic incentives encourage efficient and cost-effective consumer behavior and the development of more efficient pollution abatement techniques.
The source of the basic economic principles of environmental policy is in the theory of externalities. An attempt to provide a comprehensive description of the state of externalities theory is beyond the scope of this article, yet the policy implications of the theory are straightforward. Polluting agents are confronted with a "price" equal to the marginal external cost of their polluting activities to induce them to internalize at the margin the full social cost of their activity. Such price incentives can take the form of the well-known "Pigouvian tax," a levy on the polluting agent equal to the marginal social damage it causes.
The Vehicle Ownership Tax
Within that framework, there are three tax policy options. The first is to incorporate the "polluter pays" principle into the vehicle ownership tax. In Mexico, the vehicle ownership tax declines with the age of the vehicle-cars older than 10 years do not pay it. Raising the relative tax on the older and more polluting vehicles could stimulate vehicle owners to switch older vehicles for newer, cleaner models. The big political problem with that option is that the increased tax is regressive; those in the lower income strata tend to own older vehicles.
The Gas Tax
A second policy option is to raise the price of gasoline in order to reduce fuel consumption, thus discouraging vehicle use enough to meet the air quality standards. However, the taxing authority will find itself in the uncomfortable position of having to adjust and readjust the tax to ensure that the air standard is attained. After that, the tax has to be changed regularly to reflect the seasonal variations in the dispersive capacity of the airshed and longer-term variations due to population and income growth.
Taxing an input is a surrogate for taxing a polluting agent or activity itself when quantification of the actual pollutant emission is not feasible. Yet auto emissions are dependent upon variables such as fuel properties, combustion technology, and automotive upkeep. Taxing gasoline would therefore send an inappropriate signal, since the tax per unit of fuel would be the same whether particulate emission per unit was low or high, and could not be reasonably efficient without adjusting for weather conditions. Yet gasoline prices, set by the state-owned monopoly, are a significant part of the consumer price index and cannot, for political reasons, be adjusted frequently.
In addition, the gasoline tax would stimulate increasing suburbanization, as people would relocate to areas just outside of the reach of the taxing authority. Much of the reduction in fuel use could well be offset by commuter migration and attendant increases in vehicular miles driven per unit.
Electronic Road Pricing
The third tax option is an effluent tax on car emissions by electronic road pricing (ERP). Although that tax system is also subject to the process of trial and error in setting the appropriate tax level, developments in monitoring instrumentation, data processing, and mathematical methods of control engineering have led to unprecedented improvements in real-time air pollution forecasting that have made it possible to set the appropriate fee faster and without undesirable effects.
Under the ERP option, the use of highways or main streets is taxed in order to reduce the level of congestion. Congestion charges make drivers aware of the costs they impose on other drivers. In addition, the policy could also be used to prevent or reduce other related forms of environmental damage. The tax could be set high enough to discourage the vehicles' use, thereby facilitating attainment of the air quality standards; it would also allow regulators to take into account peak-hour periods, seasons, and the relative emissions levels of various vehicle types. In addition, on the basis of a voluntary periodic motor inspection, the car could be given an emissions rating that represents a surcharge over the basic air pollution fee. That rating could be made valid for a certain period of time or certain number of kilometers driven within the area.
Since deterioration can be assumed in the absence of renewed inspection, that surcharge would be directly related to the time elapsed since the last tune-up. Under such a system, every driver would have an incentive to have frequent tune-ups in order to qualify for a low fee.
In a 21-month pilot test, from July 1983 to March 1985 in Hong Kong, ERP demonstrated its technical feasibility. The Hong Kong program used a method of automatic vehicle identification in which each vehicle had an electronic number plate. Whenever a vehicle passed over a toll site, scanners in the road identified the vehicle number plate as it passed and relayed the vehicle identification code to a main computer. The system sent a monthly bill to the driver. The Hong Kong system performed brilliantly. Average speed went up 10 percent, fuel savings increased by 9 percent, and emissions reduction improved by 17 percent.
Given the political uncertainties accompanying Hong Kong's return to China in 1997, it is not clear whether city-wide road pricing ultimately will be implemented there. However, the ERP test pointed the way for Hong Kong to solve its huge congestion problem. It also suggests the method by which Mexico City could solve its own traffic externality problems of congestion and air pollution.
ERP has important income distribution implications, given the fact that the low-income sector uses public transportation more than high-income groups. Toll revenues could be used to finance expansion and improvement of public transportation. The high-income sector would benefit as well, because the value of travel time is proportional to the wage rate, and if time savings are proportionate for everyone, benefits will rise with income.
Marketable Emissions Permits
Marketable emissions permits are, in principle, a fully viable alternative to taxes. Instead of setting the proper tax and obtaining the efficient quantity of pollution, the regulating authority could issue emissions permits equal in the aggregate to the efficient quantity and allow consumers to bid for them.
The regulating authority could set either the price or the quantity and achieve the desired result, but when it has imperfect information concerning the marginal costs and benefits, direct control over quantity is to be preferred, since the standard itself is prescribed in quantity terms. The academic literature tends to concentrate on marketable permits for stationary pollution sources, but the framework could be applied to mobile sources as well. The objective of the marketable permits system is to facilitate allocation of the permits to their most highly valued uses, given the limited dispersive capacity of the local airshed.
The regulatory authority would have to calculate the maximum number of use-days that would not violate the air quality standard, taking into account the age and technological composition of the fleet, as well as seasonal variability. However, in the effort to apply the concept of marketable permits to mobile sources, the regulating authority will lose ground on the cost side because of the complexity of the monitoring effort, given that use-days are not the only relevant variable.
The amount of permits in the market would have to be less than the absolute ceiling of use-days in order to allow for factors such as immigration, episodic control, and temporary sales. For the proper functioning of the program there also would need to be a secondary market in which permits could be purchased. The permit unit would probably be defined as one drive-day to use on any weekday. For example, one consumer with one vehicle could purchase five permits for his vehicle in order to drive all weekdays (weekends are excluded). The permits held would be registered with the vehicle plate number. The consumer with more than one car could buy several drive-day permits for other family members.
In contrast with the passenger vehicle sector, the availability of substitutes for the use of commercial/industrial vehicles is more limited, and emissions factors vary significantly among those vehicles. Therefore, a numeraire would need to be established, and might be expressed as a light duty vehicle equivalent.
Another permit market for those types of vehicles would be necessary. The permit supply for such a market would be determined by the contribution of commercial/industrial vehicle emissions to the total vehicular emissions allowed to meet the air standards. The initial permit distribution among firms could be free and based according to their respective fleet sizes. Firms, public offices, and other institutions also could be eligible to get permit allotments based on size of payroll, tax status, or other valid proof or they could bid for permits in order to provide them to some of their employees. Given that the number of permits held is linked with the license plate number, it would be easy for firms to find out the permit status of their beneficiaries and avoid unnecessary transfers. A private, computerized permit sales network or clearing house of permits would also be necessary, with an easily accessible network of terminals where any driver could buy and sell permits for a small service fee.
ERP's Superiority
From this description, the system seems functionally complex and labor-intensive. However, it could be operated using the technology that the ERP option employs; electronic sensors could be used to determine whether the number of drive-days corresponds with the number of permits the driver holds. Awkward, unwieldy methods such as coded decals or large display cards would no longer be necessary.
Under ERP, a separate permit market for commercial/industrial vehicles would also be unnecessary. Also, regulators would be able to circumvent the problems of differentiating among passenger vehicles according their relative emissions. The only cost-effective, permit-based alternative to ERP would be a system under which the allocation is made among vehicle manufacturers in order to comply with the aggregate emissions level.
Again, the concept of tradable permits only focuses on one variable, in this case on the relative pollution abatement equipment in the vehicle fleet sales. Thus, the tradable permits option represents only a partial solution with effects in the long run, and it does not deliver efficient price formation. The ERP, on the other hand, makes the whole concept of marketable permits for mobile sources irrelevant, given the multi-variable urban environment in which the pollution from mobile sources takes place.
Even though the bulk of emissions in Mexico City comes from mobile sources, it is industry and commerce that often have to curtail activities in times of pollution alerts. The failure of the NDD program has forced more of the adjustment onto that sector, increasing the cost to the economy.
In order to adopt cost-effective policies, the extensive body of literature about marketable pollution permits for stationary sources should serve as the guideline for industrial pollution abatement policy, in conjunction with the elimination of hidden subsidies to Mexico City's industrial sector.
Cost-effectiveness is the key reason for choosing ERP over tradable permits for mobile sources, but not the only one. Other considerations that operate in ERP's favor include information requirements, monitoring and enforcement capability by government and nongovernmental organizations, and political feasibility. The ERP approach is superior because its transaction and enforcement costs would be lower than those of a tradable permits regime. The problem is serious enough to demand the full use of all available technologies in an efficient way.
Regulation is published four times a year by the Cato Institute. Editorial and business offices are located at 1000 Massachusetts Avenue, N.W., Washington, D.C., 20001. For subscription information, please write to Circulation Department, Cato Institute, same address, or call (202) 842-0200.
| Regulation | Home | Order Regulation | Publications | Search |