Practical implementations of road congestion pricing are found almost exclusively in urban areas, because
traffic congestion is common in and around city centers. Congestion pricing can be fixed (the same at all times of day and days of the week), variable (set in advance to be higher at typically high-traffic times), or dynamic (varying according to actual conditions). As congestion pricing has been increasing worldwide, the schemes implemented have been classified into four different types: cordon area around a city center; area wide congestion pricing; city center toll ring; and corridor or single facility congestion pricing.
Cordon area and area wide Cordon area congestion pricing is a fee or tax paid by users to enter a restricted area, usually within a city center, as part of a
demand management strategy to relieve traffic congestion within that area. The economic rationale for this pricing scheme is based on the
externalities or
social costs of road transport, such as
air pollution, noise,
traffic accidents, environmental and urban deterioration, and the extra costs and delays imposed by
traffic congestion upon other drivers when additional users enter a congested road. The first implementation of such a scheme was
Singapore Area Licensing Scheme in 1975, together with a comprehensive package of
road pricing measures, stringent car ownership rules and improvements in mass transit. Thanks to technological advances in
electronic toll collection, electronic detection, and video surveillance technology, collecting congestion fees has become easier.
Singapore upgraded its system in 1998, and similar pricing schemes were implemented in
Rome in 2001,
London in 2003 with extensions in 2007;
Stockholm in 2006, as a seven-month trial, and then on a permanent basis. In January 2008
Milan began a one-year trial program called
Ecopass, charging low emission standard vehicles and exempting cleaner and
alternative fuel vehicles. The Ecopass program was extended until December 31, 2011, and on January 16, 2012, was replaced by
Area C, a trial program that converted the scheme from a pollution-charge to a congestion charge. The
Gothenburg congestion tax was implemented in January 2013 and it was modeled after the Stockholm scheme. Singapore and Stockholm charge a congestion fee every time a user crosses the cordon area, while London charges a daily fee for any vehicle driving in a public road within the congestion charge zone, regardless of how many times the user crosses the cordon. Stockholm has put a cap on the maximum daily tax, while in Singapore the charge is based on a pay-as-you-use principle, and rates are set based on traffic conditions at the pricing points, and reviewed on a quarterly basis. Through this policy, the
Land Transport Authority (LTA) reports that the electronic road pricing "has been effective in maintaining an optimal speed range of 45 to 65 km/h for expressways and 20 to 30 km/h for arterial roads".
Singapore scheme In an effort to improve the pricing mechanism, and, to introduce real-time
variable pricing, Singapore's
LTA together with
IBM, ran a pilot from December 2006 to April 2007, with a
traffic estimation and prediction tool (TrEPS), which uses historical traffic data and real-time feeds with flow conditions from several sources, in order to predict the levels of congestion up to an hour in advance. By accurately estimating prevailing and emerging traffic conditions, this technology is expected to allow variable pricing, together with improved overall traffic management, including the provision of information in advance to alert drivers about conditions ahead, and the prices being charged at that moment. In 2010 the Land Transport Authority began exploring the potential of
Global Navigation Satellite System as a technological option for a second generation ERP. LTA objective is to evaluate if the latest technologies available in the market today are accurate and effective enough for use as a congestion charging tool, especially taking into consideration the dense urban environment in Singapore. Implementation of such system is not expected in the short term.
London A proposal by former Mayor of London
Ken Livingstone would have resulted in a new pricing structure based on potential CO2 emission rates by October 2008. Livingstone's successor as Mayor of London,
Boris Johnson, announced in July 2008 that the new CO2 charging structure will no longer be implemented. Johnson decided to remove the 2007 Western Extension from the congestion charging zone beginning on January 4, 2011, to increase the basic charge to , and also to introduce an automated payment system called Congestion Charging Auto Pay (CC Auto Pay), which will charge vehicles based on the number of charging days a vehicle travels within the charging zone each month, and the drivers of these vehicles will pay a reduced daily charge. In November 2012
Transport for London (TfL) presented a proposal to abolish the Greener Vehicle Discount, and the Ultra Low Emission Discount (ULED) went into effect on 1 July 2013, limiting the free access to the congestion charge zone to selected vehicles. There has been criticism because during the first ten years since the scheme was implemented, gross revenue reached about £2.6 billion, but only £1.2 billion has been invested, meaning that 54% of gross revenues have been spent in operating the system and administrative expenses. A new toxicity charge, known as T-charge was introduced from 23 October 2017. Older and more polluting cars and vans that do not meet
Euro 4 standards will have to pay an extra £10 charge within the Congestion Charge Zone (CCZ). On 8 April 2019, the T-charge was expanded into the
Ultra Low Emission Zone (ULEZ).
Milan The Ecopass pollution charge ended on December 31, 2011, and was replaced by the
Area C scheme, which went into effect on January 16, 2012, initially as an 18-month pilot program. The Area C scheme is a conventional congestion pricing scheme and is based on the same Ecopass geographic area. Vehicles entering the charging zone incur a charge of regardless of their pollution level. However, residents inside the area have 40 free entries per year and then a discounted charge of . The scheme was made permanent in March 2013. All net earnings from Area C are invested to promote
sustainable mobility and policies to reduce
air pollution, including the redevelopment, protection and development of
public transport, "soft mobility" (pedestrians,
cycling,
Zone 30) and systems to rationalize the distribution of goods.
Stockholm 's first electronic gantry at
Lilla Essingen On 1 January 2016, congestion taxes were increased in the inner-city parts of
Stockholm, and also the congestion tax was introduced on
Essingeleden motorway. This was the first increase of the tax since it was introduced permanently in 2007. The congestion tax is being introduced at the access and exit ramps of two
interchanges on Essingeleden in order to reduce traffic jams in peak periods, and with shorter traffic jams on Essingeleden, the surrounding roads are expected to have shorter tailbacks. The transport agencies involved expected to reduce traffic on Essingeleden by some 10% in peak hours. As the Stockholm congestion tax varies by time of the day, the highest increase took place at the two busiest rush hour periods, 7:30 to 8:29, and 16:00 to 17:29, from SEK 20 to SEK 30. The objective was to steer the traffic towards other times of the day and public transport, and in this way reduce congestion in the Inner City area. Also the maximum amount levied was raised to SEK 105 per day and vehicle. In 2017 New York Governor
Andrew Cuomo declared a
state of emergency for the MTA. As a solution, Cuomo proposed using congestion pricing as a means of funding the agency. Later that year, a commission to investigate the feasibility of congestion pricing found that such a scheme could benefit New York City. State lawmakers approved a congestion pricing plan in March 2019. The
Federal Highway Administration gave its final approval on June 26, 2023, allowing the MTA to begin setting toll rates for the proposed congestion zone. Implementation was scheduled for 30 June 2024, but in an announcement by Governor Kathy Hochul on 5 June 2024, the plan was indefinitely postponed. In November 2024, Hochul announced an intent for the toll to go forward with a planned implementation on 5 January 2025, at a reduced rate.
Old town centres Around
Europe several relatively small cities, such as
Durham, England;
Znojmo, Czech Republic;
Riga, Latvia; and
Valletta, Malta, have implemented congestion pricing to reduce traffic crowding, parking problems and pollution, particularly during the peak tourism season.
Durham introduced charges in October 2002, reducing vehicle traffic by 85% after a year; prior to this 3,000 daily vehicles had shared the streets with 17,000 pedestrians.
Valletta has reduced daily vehicles entering the city from 10,000 to 7,900; making 400 readily available parking places in the center. There has been a 60% drop in car stays by non-residents of more than eight hours, but there has been a marked increase of 34% in non-residential cars visiting the city for an hour or less.
Rejected proposals highlighting area of the rejected congestion charging scheme
Hong Kong conducted a pilot test on an
electronic congestion pricing system between 1983 and 1985 with positive results. However, public opposition against this policy stalled its permanent implementation. In 2002
Edinburgh, United Kingdom, initiated an implementation process; a
referendum was conducted in 2005, with a majority of 74.4% rejecting the proposal. Councils from across the
West Midlands in the United Kingdom, including
Birmingham and
Coventry, rejected the idea of imposing congestion pricing schemes on the area in 2008, despite promises from central government of transport project funding in exchange for the implementation of a road pricing pilot scheme. In 2007,
New York City shelved a proposal for a three-year pilot program for implementation in
Manhattan, and a new proposition was denied in 2008, with potential federal grants of being reallocated to other American cities.
Greater Manchester, United Kingdom, was considering a scheme with two cordons, one covering the main urban core of the
Greater Manchester Urban Area and another covering the
Manchester city centre. The measure was supported by the government, but three local authorities rejected it (
Bury,
Trafford and
Stockport); the support of two-thirds of Manchester's 10 local councils was needed for it to be implemented. A comprehensive transport investment package for Manchester, which included the congestion pricing element, was released for further public consultation and was to be subject of a referendum in December 2008. On 12 December 2008 the scheme was overwhelmingly rejected by 10 out of 10 councils by a public referendum.
Current proposals through the
Golden Gate Bridge.
United States In August 2007, the
United States Department of Transportation selected five metropolitan areas to initiate congestion pricing demonstration projects under the
Urban Partnerships Congestion Initiative, for US$1 billion of federal funding. The five projects under this initiative are
Golden Gate Bridge in
San Francisco,
State Route 520 serving downtown
Seattle and communities to its east, Interstate 95 between
Miami and
Ft. Lauderdale, Interstate 35W serving downtown
Minneapolis, and a variable rate parking meter system in
Chicago plus
Metro ExpressLanes in Los Angeles County, which replaced New York City after it left the program in 2008.
San Francisco transport authorities began a feasibility study in 2006 to evaluate the introduction of congestion pricing. The charge would be combined with other traffic reduction implementations, allowing money to be raised for public transit improvements and bike and pedestrian enhancements. The initial pricing scenarios were presented in public meetings conducted in December 2008, and the final study results were announced in November 2010, proposing modified alternatives based on the public's feedbacks, and the updated proposal calls for implementing a six-month to one-year trial in 2015.
China . Motor vehicle emissions account for 31% of the city's smog sources. The measure was initially proposed in 2010 and was recommended by the
World Bank. A similar scheme was proposed for the city
Guangzhou,
Guangdong province, in early 2010. The city opened a public discussion on whether to introduce congestion charges. An online survey conducted by two local news outlets found that 84.4% of respondents opposed the charges. Also a vehicle quota system was introduced in 2011, awarding new car licenses through a lottery, with a ceiling of 6 million units set by the city authority for 2017. In May 2016, the Beijing city legislature announced it will consider to start levying traffic congestion charges by 2020 as part of a package of measures to reform the vehicle quota system.
Brazil , near downtown
São Paulo. According to
Time magazine, São Paulo has the world's worst traffic jams. Drivers are informed through
variable message signs the prevailing
queue length. In January 2012, the
federal government of Brazil enacted the Urban Mobility Law that authorizes municipalities to implement congestion pricing to reduce traffic flows. The law also seeks to encourage the use of public transportation and reduce air pollution. According to the law, revenues from congestion charges should be destined exclusively to urban infrastructure for public transportation and non-motorized modes of locomotion (such as
walking and
cycling), and to finance public subsidies to transit fares. The law went into effect in April 2013. In April 2012, one of the committees of the
São Paulo city council approved a bill to introduce congestion pricing within the same area as the existing
road space rationing () by the last digit of the license plate, which has been in force 1996. The proposed charge is R$4 (~ ) per day and it is expected to reduce traffic by 30% and raise about R$2.5 billion (~ billion) per year, most of which will be destined to the expansion of the
São Paulo Metro system and bus corridors. The bill still needs approval by two other committees before going for a final vote at the city council. Since 1995, 11 bills have been presented in the city council to introduce congestion pricing. Opinion surveys have shown that the initiative is highly unpopular. A survey by
Veja magazine found that 80% of drivers are against congestion pricing, and another survey by
Exame magazine found that only 1% of São Paulo's residents support the initiative, while 30% find that extending the metro system is a better solution to reduce traffic congestion. São Paulo's strategic urban development plan "SP 2040", approved in November 2012, proposes the implementation of congestion pricing by 2025, when the density of metro and bus corridors is expected to reach 1.25 km/km2. The Plan also requires ample consultation and even a
referendum before beginning implementation.
Thailand In October 2024, Thailand's
Ministry of Transport announced plans for a 40-50
Baht congestion charge for motorists who enter streets in inner
Bangkok. The funds would be used to subsidize a 20 Baht fare for all railway lines in
Greater Bangkok.
Urban corridors and toll rings " Freeway, crossing downtown with 100% free flow,
Santiago, Chile Congestion pricing has also been implemented in urban freeways. Between 2004 and 2005,
Santiago implemented the first 100% non-stop urban toll for a freeway passing through a downtown area, charging by the distance traveled. Congestion pricing has been used since 2007 during rush hours in order to maintain reasonable speeds within the city core.
Norway pioneered the implementation of
electronic urban tolling in the main corridors of Norway's three major cities:
Bergen (1986),
Oslo (1990), and
Trondheim (1991). In Bergen cars can only enter the central area using a toll road, so that the effect is similar to a congestion charge. Though initially intended only to raise revenues to finance road infrastructure, the urban toll ring at Oslo created an unintended congestion pricing effect, as traffic decreased by around 5%. The
Trondheim Toll Scheme also has congestion pricing effects, as charges vary by time of day. The Norwegian authorities pursued authorization to implement congestion charges in cities, and legislation was approved by Parliament in 2001. In October 2011 the Norwegian government announced the introduction of rules allowing congestion charging in cities. The measure is intended to cut
greenhouse gas and air pollutant emissions, and relieve traffic congestion. , Norwegian authorities have implemented urban charging schemes that operates both on the motorways and for access into downtown areas in five additional cities or municipalities:
Haugesund,
Kristiansand,
Namsos,
Stavanger, and
Tønsberg. The Norwegian electronic toll collection system is called
AutoPASS and is part of the joint venture
EasyGo.
Single facilities Urban HOT lanes at
91 Express Lanes, at
Orange County, California Congestion pricing has also been applied to specific roadways. The first of this kind of specific schemes allowed users of low or single-occupancy vehicles to use a
high-occupancy vehicle lanes (HOV) if they pay a toll. This scheme is known as
high-occupancy toll lanes (HOT) lanes, and it has been introduced mainly in the United States and
Canada. The first practical implementations was
California's private toll
91 Express Lanes, in
Orange County in 1995, followed in 1996 by
Interstate 15 in
San Diego. There has been controversy over this concept, and HOT schemes have been called "
Lexus" lanes, as critics see this new pricing scheme as a perk to the rich. According to the
Texas A&M Transportation Institute, by 2012 there were in the United States 722 corridor-miles of HOV lanes, 294 corridor-miles of HOT/Express lanes and 163 corridor-miles of HOT/Express lanes under construction. Congestion pricing in the form of variable tolls by time-of-the-day have also been implemented in bridges and tunnels providing access to the
central business districts of several major cities. In most cases there was a toll already in existence.
Dynamic pricing is relatively rare compared to
variable pricing. One example of dynamic tolling is
Interstate 66 in the Washington, D.C., metro area, where at times of severe congestion tolls can reach almost . However, on average, round trip prices are much lower: $11.88 (2019), $5.04 (2020), $4.75 (2021). in January 2009. In March 2001, the
Port Authority of New York and New Jersey (PANYNJ) implemented a discount on regular toll fees during off-peak hours for those vehicles paying electronically with an
E-ZPass issued in New York State. These discount toll was implemented at several tunnels and bridges connecting
New York City and
New Jersey, including the
George Washington Bridge,
Lincoln Tunnel, and
Holland Tunnel, and at some other bridges administered by PANYNJ. Since March 2008, qualified low-emission automobiles with a
fuel economy of at least 45 miles per gallon are eligible to receive a Port Authority Green Pass, which allows for a 50% discount during off-peak hours as compared to the regular full toll. In January 2009, variable tolls were implemented at
Sydney Harbour Bridge, two weeks after upgrading to 100% free-flow
electronic toll collection. The highest fees are charged during the morning and afternoon peak periods; a toll 25% lower applies for the shoulder periods; and a toll lower than the previously existing is charged at nights, weekends, and public holidays. This is
Australia's first road congestion pricing scheme, and has had only a very minor effect on traffic levels, reducing them by 0.19%. In July 2010 congestion tolls were implemented at the
San Francisco–Oakland Bay Bridge. The Bay Bridge congestion pricing scheme charges a toll from 5 a.m. to 10 a.m. and 3 p.m. to 7 p.m., Monday through Friday. During weekends cars pay . The toll remained at the previous toll of at all other times on weekdays. According to the
Bay Area Toll Authority fewer users are driving during the peak hours and more vehicles are crossing the Bay Bridge before and after the 5–10 a.m. period in which the congestion toll goes into effect. The agency also reported that commute delays in the first six months have dropped by an average of 15 percent compared with 2009. When the congestion tolls were proposed, the agency expected the scheme to produce a 20 to 30 percent drop in commute traffic.
Non-urban Autoroute A1 in Northern
France is one of the few cases of congestion pricing implemented outside of urban areas. This is an expressway connecting
Paris to
Lille, and since 1992 congestion prices have been applied during weekends with the objective of spreading demand on the trip back to Paris on Sunday afternoons and evenings.
Research Measurement of effects In a road network, congestion can be considered a specific measure of the time delay in a journey or time lost through
traffic jams. Delays can be caused by some combination of traffic density, road capacity, and the delaying effects of other road users and traffic management schemes such as traffic lights, junctions, and street works. This can be measured as the extra journey time needed to traverse a congested route when compared to the same route with no such interference. However, this technical definition of congestion as a measurement of delay can get confused and used interchangeably with traffic density in the public mind. To measure the true effects of any traffic management scheme it is normally necessary to establish a baseline, or "do nothing" case, which estimates the effects on the network without any changes other than normal trends and expected local changes. Notably this was not done for the
London congestion charging scheme, which has led to claims that it is not possible to determine the extent of the actual influence of the scheme. Regardless of the scheme's impact, in a retrospective analysis
Transport for London (TfL) estimated there would have already been a significant reduction in traffic as a consequence of parking policies and increased congestion due to traffic management and other interventions that had the effect of reducing highway capacity. In 2006, the last year before the zone was expanded, TfL observed that traffic flows were lower than in any recent year, while network traffic speeds were also lower than in any recent year. In 2013, ten years since its implementation, TfL reported that the congestion charging scheme resulted in a 10% reduction in traffic volumes from baseline conditions, and an overall reduction of 11% in vehicle kilometres in London between 2000 and 2012. Despite these gains, traffic speeds have also been getting progressively slower over the past decade, particularly in central London. TfL explains that the historic decline in traffic speeds is most likely due to interventions that have reduced the effective capacity of the road network in order to improve the urban environment, increase road safety and prioritise public transport, pedestrian and cycle traffic, as well as an increase in road works by utilities and general development activity since 2006. TfL concludes that while levels of congestion in central London are close to pre-charging levels, the effectiveness of the congestion charge in reducing traffic volumes means that conditions would be worse without the Congestion Charging scheme.
New York City Congestion Pricing Environmental Assessment Study In May 2023, the Metropolitan Transportation Authority (MTA) finalized and published the Environmental Assessment (EA) of the congestion program which included a public comment period of 30 days, ending on June 12, 2023. The EA included a comprehensive regional study of 22 million people taking 28.8 million journeys per average weekday in a 28-county area covering New York, New Jersey and Connecticut. True social-cost pricing of metropolitan travel has proven to be a theoretical ideal that so far has eluded real-world implementation. The primary obstacle is that except for professors of transportation economics and a cadre of vocal environmentalists, few people are in favor of considerably higher charges for peak-period travel. Middle-class motorists often complain they already pay too much in gasoline taxes and registration fees to drive their cars, and that to pay more during congested periods would add insult to injury. In the United States, few politicians are willing to champion the cause of congestion pricing for fear of reprisal from their constituents. Critics also argue that charging more to drive is elitist policy, pricing the poor off of roads so that the wealthy can move about unencumbered. It is for all these reasons that peak-period pricing remains a pipe dream in the minds of many. Both Button and Small et al., have identified the following issues: • The real-world demand functions for urban road travel are more complex than the theoretical functions used in transport economics analysis. Congestion pricing was developed as a first-best solution, based on the assumption that the optimal price of road space equals the marginal cost price if all other goods in the economy are also marginal cost priced. In the real world this is not true, thus, actual implementation of congestion pricing is just a proxy or second-best solution. Based on the economic principles behind congestion pricing, the optimal congestion charge should make up for the difference between the average cost paid by the driver and the marginal cost imposed on other drivers (such as extra delay) and on society as a whole (such as air pollution). The practical challenge of setting optimal link-based tolls is daunting given that neither the demand functions nor the link-specific speed-flow curves can be known precisely. Therefore, transport economists recognize that in practice setting the right price for the congestion charge becomes a trial and error experience. • Inequality issue: A main concern is the possibility of undesirable distribution repercussions because of the diversity of road users. The use of the tolled road depends on the user's level of income. Where some cannot afford to pay the congestion charge, then this policy is likely to privilege the middle-class and rich. The users who shift to some less-preferred alternative are also worse off. The less wealthy are the more likely to switch to public transit.
Road space rationing is another strategy generally viewed as more equitable than congestion pricing. However, high-income users can always avoid the travel restrictions by owning a second car and users with relatively inelastic demand (such as a worker who needs to transport tools to a job site) are relatively more impacted. • There are difficulties in deciding how to allocate the revenues raised. This is a controversial issue among scholars. The revenues can be used to improve public transport (as is the case in London), or to invest in new road infrastructure (as in Oslo). Some academics make the case that revenues should be disposed as a direct transfer payments to former road users. Congestion pricing is not intended to increase public revenues or to become just another tax, however this is precisely one of the main concerns of road users and taxpayers. One alternative, aimed at avoiding inequality and revenue allocation issues, is to implement a
rationing of peak period travel through mobility rights or revenue-neutral credit-based congestion pricing. This system would be similar to the existing
emissions trading of
carbon credit. Metropolitan area or city residents, or the taxpayers, would be issued mobility rights or congestion credits, and would have the option of using these for themselves, or trading or selling them to anyone willing to continue traveling by automobile beyond their personal quota. This trading system would allow direct benefits to be accrued by those users shifting to public transportation or by those reducing their peak-hour travel rather than the government.
Public controversy Experience from the few cities where congestion pricing has been implemented shows that social and political acceptability is key. Public discontent with congestion pricing, or rejection of congestion pricing proposals, is due mainly to the inequality issues, the economic burden on neighboring communities, the effect on retail businesses and the economic activity in general, and the fears that the revenues will become just another tax. Congestion pricing remains highly controversial with the public both before and after implementation. This has in part been resolved through
referendums, such as after the seven-month trial period in
Stockholm; however this creates a debate as to where the border line for the referendum should go, since it is often the people living outside the urban area who have to pay the tax, while the external benefit is granted to those who live within the area. In Stockholm there was a majority in the referendum within the city border (where the votes counted), but not outside. Some concerns have also been expressed regarding the effects of cordon area congestion pricing on economic activity and land use, as the benefits are usually evaluated from the urban transportation perspective only. However, congestion pricing schemes have been used with the main objective of improving urban quality and to preserve historical heritage in the small cities. while others show that businesses were then supporting the charge six months after implementation. Reports show business activity within the charge zone had been higher in both
productivity and
profitability and that the charge had a "broadly neutral impact" on the London wide economy, while others claim an average drop in business of 25% following the 2007 extension. Other criticism has been raised concerning the environmental effects on neighborhoods bordering the congestion zone, with critics claiming that congestion pricing would create "parking lots" and add more traffic and pollution to those neighborhoods, and the imposition of a
regressive tax on some commuters. Stockholm's trial of congestion pricing, however, showed a reduction in traffic in areas outside the congestion zone. Other opponents argue that the pricing could become a tax on middle- and lower-class residents, since those citizens would be affected the most financially. The installation of cameras for tracking purposes may also raise civil liberties concerns.
Effects A 2019 study of congestion pricing in Stockholm between 2006 and 2010 found that in the absence of congestion pricing that Stockholm's air would have been 5 to 15 percent more polluted between 2006 and 2010", and that young children would have suffered substantially more asthma attacks. A 2020 study that analyzed driving restrictions in Beijing estimated that the implementation of congestion pricing would reduce total traffic, increase traffic speed, reduce pollution, reduce
greenhouse gas emissions, reduce traffic accidents, and increase tax revenues. A 2020 study of London found that
congestion pricing (introduced in 2003) led to reductions in pollution and reductions in driving, but it increased pollution from diesel vehicles (which were exempt from the congestion pricing). A 2021 study found that congestion pricing reduced emissions through downsizing commuting distances and housing sizes. A 2013 study found that after congestion pricing was implemented in Seattle, drivers reported greater satisfaction with the routes covered by congestion pricing and reported lower stress. A 2016 study found that more people used public transportation due to increases in congestion pricing in Singapore. A 2016 study found that the prices of retail, but not office and residential, real estate dropped by 19% within the cordoned-off areas of Singapore where congestion pricing was in place relative to the areas outside of the area. == Waterways ==