The last three decades of the 20th century brought significant improvements in air quality to cities in the United States. Between 1970 and 2002, emissions of nitrogen dioxide, ozone, sulfur dioxide, particulate matter, carbon monoxide, and lead decreased by an average of 48 percent. This has not been the case in Mexico City where automobiles are responsible for the vast majority of emissions. As of the time of this study, levels of major air pollutants in Mexico City routinely exceeded maximum exposure limits established by the World Health Organization (WHO). For example, in 2005 the WHO warned that 8-hour average ozone levels exceeding 100 micrograms per cubic meter threaten human health. Between 1986 and 2005, this guideline was exceeded in Mexico City for 92 percent of all days during that time period.
Airborne pollutants are linked to respiratory infections, chronic respiratory illness, and aggravation of existing cardiovascular disease. There are also studies that link high levels of air pollution to infant mortality. The social costs of air pollution are considerably higher when people change their behavior to reduce exposure. In essence, air pollution is associated with high social costs such as health issues or behavior changes, but the cost of polluting is comparably low. This combination is worrisome as trends in population and vehicle growth throughout the developing world threaten to worsen pollution levels and the social costs associated with them.
In November 1989, record levels of ozone and other airborne pollutants led to the establishment of driving restriction program called Hoy No Circula (HNC) throughout Mexico City’s metropolitan area. HNC is still in place today. The program bans driving of private vehicles one weekday per week between 5am and 10pm based on the last digit of the vehicle’s license plate. When implemented in 1989, the program applied to 2.3 million vehicles or 460,000 vehicles per day. Enforcement of the program is easy, inexpensive and required small public investment prior to being imposed. Because enforcement is tough and fines and other costs of violating driving rules are high, compliance is near universal. Similar programs have been implemented in Bogotá, Colombia, Santiago, Chile and São Paolo, Brazil.
This study measures the effect of HNC on air quality using hourly air pollution records from monitoring stations recorded by the Automated Environmental Monitoring Network of the city environmental agency. The network was established in 1986 and reports hourly measures of carbon monoxide, nitrogen dioxide, ozone, nitrogen oxides, and sulfur dioxide. The measures are then reported to the public in the form of the Metropolitan Air Quality Index.
The empirical analysis focuses on the period 1986 through 1993—an eight year window around the implementation of the program. The data sample is restricted to the observations from stations that were operating in 1986: 15 stations for carbon monoxide and sulfur dioxide, nine stations for ozone, and five stations for nitrogen dioxide and nitrogen oxides. Pollution levels for the five pollutants tracked by the monitoring stations are compared before and after the restrictions, with levels in previous years as the comparison group. The analysis controls for confounding factors by restricting the sample to a relatively narrow time window around the implementation of HNC and by using a regression discontinuity design. The study also considers a number of additional specifications such as gasoline sales, vehicle registration, public transportation ridership, and taxi use.
Results and Policy Implications
Across all pollutants and specifications there is no evidence that the program improved air quality both in terms of average pollution levels or daily peak levels. Evidence from the study suggests that the program prompted a relative increase in air pollution during weekends and weekday late hours; however there is no absolute improvement in air quality during any period for any pollutant. It was hoped that the driving restrictions would cause drivers to substitute to low-emissions forms of transportation but consistent with findings on air quality, HNC did not reduce the consumption of gasoline in Mexico City nor did it increase the use of public transportation. Instead, evidence from vehicle registrations and automobile sales indicate that the program led to an increase in the total number of vehicles in circulation as well as a shift in the composition of vehicles toward used, and thus higher-emitting, vehicles.
Driving restrictions attempt to address a market failure where costs of emissions are small relative to the total social costs by imposing quantity constraints on transportation goods. This analysis demonstrates how quantity restrictions are not a guarantee of improvement in air quality. This is relevant to environmental policy in Mexico City where air pollution remains a problem and demonstrates the importance of conducting pre-implementation evaluations of substitution patterns likely to be induced by programs like HNC. More generally, the analysis has implications for air quality and transportation policies throughout the urban developing world where population and vehicle growth threaten to intensify pollution.
 EPA (U.S. Environmental Protection Agency). 2003. “Latest Findings on National Air Quality: 2002 Status and Trends.” EPA 454/K-03-001. Washington, DC: U.S. EPA.
 WHO (World Health Organization). 2005. “WHO Air Quality Guidelines Global Update 2005: Report on Working Group Meeting, Bonn, Germany, 18-20 October 2005.” Worth health Org., Geneva.
 As of 2008, the program also restricts the driving of private vehicles one Saturday per month. Additional modifications to the program in 1997 and 2004 have made certain additional low-emission vehicles exempt from the restrictions and removed exemptions for some taxis and buses. Current information (in Spanish) on the program can be found at: http://www.sma.df.gob.mx/verificentros/index.php?op=inicio&opsubmenu=hoynocircula