Improving our understanding of traffic to improve our response: “It’s the economy, stupid!”
Congested cities or congested society: space, time… and us
Urban congestion, as defined by traffic experts, appears when transit demand exceeds the infrastructure capacity. This means there are two states of traffic: fluid traffic and congested traffic arising at the critical point beyond which transit demand exceeds infrastructure capacity.
Two types of phenomena can lead to transit network congestion. They are known as recurring congestion (or demand congestion) and non-recurring congestion (or offer congestion):
- In the first case, traffic jams occur when demand increases beyond the availability offered by the road network. This phenomenon notably emerges during rush hour or at times of holiday or vacation traffic.
- In the second case, traffic jams are triggered by a sudden or planned reduction in the road network capacity. These traffic episodes notably arise during construction, lane closures or traffic accidents.
A traffic jam thus occurs above all due to an imbalance between capacity and demand. In this respect, cities are particularly conducive to creating traffic jams…
For our societies to operate efficiently and effectively, it’s important to synchronize our working hours. This means that working people and students need to meet in the same places and at the same times in order to interact. This need imposed by our societies means that a majority of people travel during the same periods: 7:00-9:00 AM and 4:00-6:00 PM. In addition, the spatial concentration of jobs, exacerbated by urbanization, contributes to greater use of road networks around employment hubs. Although digital technology promised to do away with traditional office spaces and eliminate commutes, the reality is that working people continue to commute, even doing so across longer distances.
Several factors explain this growing reliance on commuting over long distances. First of all, this trend was facilitated by the presence of long-distance transportation networks (suburban trains, buses) and by the diminishing cost of car ownership. Faster and more available transportation has expanded the boundaries of cities and enabled people to live farther from downtown. In addition, the increase in travel distances has become a solution to the tensions between the job market, rising property values and disparities between regional appeal.
Far from the return to rural living enabled by computers and the Internet, society remains decidedly urban. For Edward Glaeser, Economics Professor at Harvard University and a specialist in urban growth factors, this situation comes down to the fact that cities are not just places where people work. They are also spaces of consumption where people like to spend their time.Contrary to the changes predicted by the arrival of computers and digital technology, cities have gradually expanded their sprawl just as daily commutes have stretched across longer distances and durations. People are becoming increasingly mobile. This trend has gained particular steam in Europe in recent years .In France, the distances traveled by commuters  increased by an average of 1.6 km between 1999 and 2013 ; the same trend occurred in the United Kingdom between 2001-2011. A study led by the Urban Sociology Laboratory (LaSUR) at the Swiss Federal Institute of Technology Lausanne (EPFL) indicated that in 2015, between 11-15% of European working people between 25-54 fit the description as “extremely mobile“. In France, the share of trips under 10 km decreased, just as trips between 20-50 km increased. As these distances increase and extend farther from downtown, cars remain the primary –and often the only –mode of transportation (Fig. 2), which, when not optimized, contributes to disproportional use of roads and favors the emergence of traffic jams.
This excessively high demand leads to the emergence of traffic jams when use of the road network exceeds its capacity. This excessive concentration of traffic prevents vehicles from reaching the optimal travel speed. The main cause of this phenomenon pertains to the suboptimal character of personal cars. Often singled out for their ill effects, personal cars display a low occupancy rate relative to their capacity. In Europe, vehicle occupancy rates fell between 1990 and 2005 from 1.65 to 1.45 passengers per vehicle. At the same time, the number of kilometers traveled per passenger has increased over the same period: by 45% in Germany, 28% in France and 15% in the United Kingdom. As a result: the number of vehicles on roads is growing faster than the number of people being transported.
What efforts are possible?
It is plain to everyone that road networks cannot simultaneously carry an increasing number of people who also want to travel over longer distances with more vehicles. Several types of efforts are available to remedy this problem: increase the road network supply, reduce demand or shift demand to other areas.
By defining traffic congestion as the result of car transportation demand that exceeds the road network capacity, it may seem intuitive to increase the road network capacity or build new roads to absorb traffic congestion. Although this line of reasoning may seem fool-proof, the reality is more complex on the road and so the actual results are less conclusive.
In Houston, to ease traffic in what the American Highway Users Alliance (AHUA) ranked as the second-worst bottleneck in the nation, which wasted 25 million hours of commuter time every year, the city decided to increase capacity along a stretch of the highway . At the interchange with Houston’s beltway, the Katy Freeway(Fig. 3) was expanded to become the widest highway in the world with nearly 26 lanes in 2008 (12 main lanes, 8 feeder lanes, 4-6 toll lanes). The project tripled the original capacity of the highway, which upon its construction in 1968, could carry 80,000 vehicles per day along 6 lanes . However, congestion along this route has continued to grow even after opening these additional lanes: it has since increased by 33%.
As this example shows: increasing a roadway’s capacity without changing the cost of use  attracts greater transportation demand. Since newly built or widened roads are initially less congested, if their cost does not change, they become more competitive. For that reason, traffic along these routes continues to increase until bottlenecks form once again, thus reducing its appeal. The reality of traffic means that by increasing the supply, we simultaneously create new demand . This is known as induced demand generated by increasing the road network supply. As illustrated by the example in Houston, traffic may actually become worse after building or widening a highway.
Instead of permanently increasing a roadway’s capacity, providing temporary access to an additional lane in order to prevent bottlenecks is another solution that has been tested. In 2017, in Rennes, a study undertaken through a Pacte État-Métropole agreement aimed to study the possibility of allowing some users (bus and carpool) to use the emergency lane along the Nantes-Rennes highway (RN 137) during traffic jams . Without permanently increasing the roadway’s capacity, and thus avoiding the risk of worsening the original situation, this operation should make it possible to ease traffic while encouraging new behaviors and influencing transportation demand.
Whether due to fears of worsening the initial situation or a lack of public funding, supply-side efforts are not always possible. In the absence of solutions for increasing road capacity, it may be possible to influence transportation demand. In theory, this type of effort has a direct impact on vehicle flows, traffic concentration and, subsequently, traffic congestion. How? By getting more passengers in cars, first of all. In 1997, it was estimated that the average car in Europe transported 1.1-1.2 passengers during commutes . In suburban areas, where nearly 97% of residents used their car regularly, that rate fell to 1.06 passengers . Considering this situation, the solution seemed simple. “To cut traffic jams, we simply need to reach 5% self-driving cars“; “if we achieve 1.7 [passengers],we would solve traffic jams in Paris“, read some of the headlines in the French press.
This approach aims primarily to encourage carpooling during commutes. Experiments like the one launched in September 2017 in Toulouse and Reims by BlaBlaCar aspire to develop carpooling, which according to ADEME accounted for just 3% of all commutes in 2015 . Efforts like developing roads to encourage carpooling, creating parking spaces and meeting points or including carsharing in corporate transportation plans (PDE) provide incentives for carpooling during the daily commute.
Another solution consists in desynchronizing transportation rhythms. Just like similar efforts on electric grids, this involves eliminating a portion of demand during peak hours. This can be done either temporally, by providing incentives to travel at a different time (earlier or later), or spatially, by choosing alternative routes. In any case, any such actions will require the agreement of companies and the creation of reliable alternative routes. Moreover, as with the electric grid, how much will users need to receive in return before changing their habits?
Finally, what would happen to traffic congestion if we simply commuted less?This is the question posed by remote work and the gradual emergence of shared workspaces, for example near or even inside train stations. Though a growing share of the working population is starting to adopt remote work, this solution is still very limited, while a majority of workers (60%) and a substantial portion of jobs (45%) are not eligible for remote work .
What is the right price of traffic congestion?
However, it is clear that solutions like remote work and carpooling, though they are already available and are occasionally put into place, have struggled to take off with a mass audience. At the same time, traffic congestion continues to get worse.
Faced with the long-term inefficiency of efforts focusing either on road network supply or transportation demand, other approaches of economic nature have been envisioned. According to economist Anthony Downs, the previously mentioned efforts cannot solve traffic congestion . Even worse, the nearly automatic reflex of rebalancing transportation demand can also lead some people who previously used public transit to opt instead for personal cars, due to the extra appeal generated by the decongested roadway. This is explained by what Downs calls “triple convergence“. When it comes to transport networks, traffic flows adjust automatically. For this reason, the additional space made available by efforts focusing on road network supply (widening, new construction) or on transportation demand (reducing the number of vehicles on the road) will quickly reach capacity.
Within these conditions, a single mechanism has the ability to neutralize the triple convergence phenomenon: increasing the cost of using a car by imposing a geographic fee or by increasing taxes on petroleum products like gasoline.
Moving beyond a physical approach to traffic jams by adding an economic angle: for economists like Anthony Downs, the only way to put the brakes on rising traffic congestion and stabilize it at an optimal level is to attach a cost to the negative externality it generates. However, it is still necessary to find a way to calculate this cost and evaluate the negative externalities generated by cars and their use: congestion, pollution, noise pollution, premature decay of roads, stress, anxiety, etc. Recall that a negative externality corresponds to the moment when the consumption of a good or service –in this case, road use – is affected negatively by the consumption of other individuals. Traffic congestion represents a unique type of externality : the people caught up in a traffic jam are subjected to it just as they also cause it .
The total cost of urban congestion is often calculated by totaling the costs of the various negative externalities it generates. By adding up the toll of decay, wasted time, pollution and health consequences, some experts estimated the total cost of traffic jams in France at 17 billion euros per year in 2014 , or about 0.8% of the nation’s gross domestic product (GDP). However, not everyone agrees with this calculation. Rémy Prudhomme casts doubt on these figures, which he deems scarcely credible. Overestimating the total cost of traffic congestion constitutes a risk for anyone seeking to develop the fairest possible assessment. These approaches differ based on how we define traffic volumes and the duration and value of time wasted in traffic jams. This type of calculation tends to overestimate the amount of time wasted in traffic jams by comparing it with an ideal control situation in which roads are completely empty of any cars and traffic moves without obstruction. This postulate is questionable in that roads are not designed to be unused. Therefore, this control situation is not realistic. In urban areas, roads are almost always congested .
Calculating the marginal cost of traffic congestion
In 1999, Rémy Prudhomme presented a model for calculating the cost of traffic congestion, aiming to provide a more credible method than the standard model, which was based on imprecise definitions of congestion and its cost.
First of all, he deconstructs the founding myth of this type of calculation. In his view, the cost of traffic congestion should not be calculated in relation to an extremely unrealistic scenario in which there are no cars on the road. Roads are built to be used.
As a result, Prudhomme bases his method for calculating the cost of traffic congestion on a natural traffic equilibrium (point A). This more realistic scenario comes into play when an additional driver pays a private cost (primarily composed of the value of time spent on the road and the cost of the vehicle’s operation) that is equal to the benefit the driver receives from using the road. Beyond this balance, the additional driver will see the cost of using the road exceed the value they receive from it, so they will logically decide not to use it. For Prudhomme, although this equilibrium is natural, it is still not optimal for society.
Prudhomme therefore considers a social cost, which corresponds to the private cost paid by the driver, as well as the cost that the driver’s vehicle imposes on all other vehicles when it is on the road. This second cost curve compares demand at a second equilibrium point (B), which Prudhomme considers the optimal traffic equilibrium for society. Beyond this equilibrium, it’s not just the use value for the driver that decreases, but the use value for all other drivers on the same road.
For the author, a cost occurs, and is paid for by society, when the equilibrium is natural instead of optimal. One initial conclusion derives from this demonstration. Natural equilibrium is a scenario that often already includes traffic jams because, according to Prudhomme, the natural use of a road is nearly always greater than its capacity. In this way, the goal of public mobility policies is not to eliminate all congestion, but to stabilize it at an optimal level.
For society, reaching an optimal traffic equilibrium requires different types of efforts. According to Downs, the most effective action consists in increasing the cost of using infrastructure in order to reduce demand. Enforcing a fee equates to charging the additional driver (who disturbs the balance to the natural state of traffic) for the delays they impose on other drivers.
Aside from the difficulties involved in assessing a fair price, the very concept of charging for urban traffic congestion raises several issues. In the first place, it requires making people pay to drive in specific areas at certain times of day. This type of fee is especially unpopular, because using roads was once seen as a right acquired simply by paying taxes. Enforcing an additional fee on traffic congestion amounts to charging a double tax for the use of a single space.
Next, these measures can be perceived as antilabor. In fact, though one portion of the population may be able to avoid driving or pay the fine, another portion may not have this option. Finally, drivers already pay a large portion of the cost of congestion through the time wasted in traffic jams every day. To achieve what it sets out to accomplish, traffic congestion fees will need to overcome all these setbacks.
Unlike an actual public utility, which offers limited resources, the road network is expandable. Nevertheless, once its expansion capacityis reached, either because of a lack of space to build new roads, or because of a lack of resources to do so, we can consider the road network as a finite resource and, therefore, a public utility for the portion not under concession (and therefore free of charge). On these roads, traffic jams constitute a perfect example of “the Tragedy of the Commons”. Theorized by Hardin in 1968, this phenomenon is defined as the overuse of a shared and limited resource resulting in a no-win situation for all the economic players competing for its use. In other words, everyone loses in a traffic jam. This theory is useful for thinking about how drivers can influence this situation and adopt the best decisions together .
Resolving this situation raises a complex problem: how to simultaneously coordinate the individual decisions of a large mass of people in order to regulate the use of a freely accessible consumer good? For now, each driver will tend to operate in a rational way: their top priority will be to reduce their total cost of traveling. As indicated above, the organization of our society leads all drivers to make these types of individualistic decisions at the same time, thus favoring the emergence of traffic jams and a situation in which everyone loses by getting caught in traffic jams . This is an imperfect configuration, in which everyone stuck in traffic jams stands to benefit from improving the situation.
To influence the emergence of traffic jams, it is necessary for each driver to act in a way that considers the actions of other drivers. However, choosing a route is not simply a process of opting for route A instead of route B. Transportation involves a combination of choices connecting a starting point to a destination. Highway exits, intersections, detours: every crossroads presents a new alternative. Optimizing a trip is a process that takes place at every instant, since an infinite number of itineraries are available. It is not farfetched to think that such complexity can put the rationality and good intentions of drivers to the test when it comes to solving urban traffic congestion. For this reason, individualistic behaviors within this scenario actually appear as perfectly logical.
Today, several tools enable drivers to account for the choices of their fellow drivers. For example, when they encounter a slowdown or bottleneck, drivers would previously make a choice favoring their own utility instead of the good of everyone . Now, in the case of a slowdown, digital tools theoretically make it possible to guide choices towards routes that will optimize use of the entire road network. However, as noted above, digital traffic management tools frequently worsen the initial situation. Because, while these tools help to reduce travel time for individual users, they are not yet able to account for the decisions of drivers who follow traffic app suggestions or who use another service. One identifiable solution for reducing traffic congestion may lie in connecting all traffic regulation services. This would involve greater collaboration between public and private traffic management services.
Remote working, building additional lanes, taxing traffic: between apparently short-term solutions and medium-term quandaries (building new lanes), complex transformational changes involving all players (changing social rhythms) and technically efficient but politically difficult measures (charging drivers), the strategies for reducing traffic congestion extend far beyond the realm of digital technology. As underlined by Martin Wachs, professor at UC Berkeley, “we consistently label congestion a major problem to be solved but find it unacceptable to adopt the most effective solutions“.
The real challenge, therefore, is not to do away with traffic congestion but to regulate it successfully – a challenge in which digital has a key role to play.
This article is the third installment of a series of four episodes by La Fabrique de la Cité investigating the role of digital technologies in the resolution of cities’ congestion problem.
 Expressed in number of vehicles using the infrastructure.
 Christine Buisson, Jean-Baptiste Lesort, Comprendre le trafic routier : Méthodes et calculs,CERTU, 2010.
 Richard Dowling, Alexander Skabardonis, Michael Carroll, Zhongren Wang, Methodology for Measuring Recurrent and Nonrecurrent Traffic Congestion, Transportation Research Record: Journal of the Transportation Research Board, 1867: 60–68, 2004.
 Jean Coldefy, Villes et voitures : pour une réconciliation, Fondation pour l’innovation politique, June, 2018. (Link: http://www.fondapol.org/wp-content/uploads/2018/05/127-Jean-Coldefy_2018-05-31_web.pdf)[Accessed on August 28, 2018].
 Emmanuel Ravalet, Grands mobiles ou grands sédentaires?, Forum Vies Mobiles – Préparer la transition mobilitaire, February 18, 2014. (Link: http://fr.forumviesmobiles.org/video/2014/02/18/grands-mobiles-ou-grands-sedentaires-2175] [Accessed on August 28, 2018].
 Greg Rosalsky, The Economics Of The Office: Why Do We Still Commute?, Pacific Standard, October 30, 2017. (Link: https://psmag.com/economics/why-do-we-still-commute] [Accessed on August 28, 2018].
 Fabrice Breithaupt, La grande mobilité liée au travail est un phénomène de société, Entretien avec Vincent Kaufmann, Sociologue, EPFL, La Tribune de Genève, 19 novembre 2014 [En ligne : https://transport.epfl.ch/wp-content/uploads/2018/08/page012.pdf] [Accessed on August 28, 2018].
 According to the French National Institute of Statistics and Economic Studies (INSEE), a commuter is an employed person who does not working in his home city.
 Institut national de la statistique et des études économiques, De plus en plus de personnes travaillent en dehors de leur commune de résidence, INSEE Première n°1605, June 30, 2016.
 “Extremely Mobile”: a person who devotes more than two hours a day for their commute to work on average.
 Gaële Lesteven, Les stratégies d’adaptation à la congestion automobile dans les grandes métropoles. Analyse à partir des cas de Paris, São Paulo et Mumbai, Géographie, Université Panthéon-Sorbonne – Paris I, 2012.
 European Environment Agency,Occupancy rates of passenger vehicles, December 16, 2008. [En ligne : https://www.eea.europa.eu/data-and-maps/indicators/occupancy-rates-of-passenger-vehicles/occupancy-rates-of-passenger-vehicles] (Consulté le 31 août 2018).
 Kurt Van Dender, Martin Clever, Recent Trends in Car Usage in Advanced Economies – Slower Growth Ahead?, International Transport Forum, September, 2009. [En ligne : https://www.itf-oecd.org/sites/default/files/docs/dp201309.pdf] [Accessed on August 28, 2018].
 Joe Cortright, Reducing congestion: Katy didn’t, CityCommentary, City Observatory, December 16, 2015. (Link: http://cityobservatory.org/reducing-congestion-katy-didnt/)[Accessed on August 28, 2018].
 Federal Highway Administration, Project Profile: Katy Freeway Reconstruction, 2008. (Link: https://www.fhwa.dot.gov/ipd/project_profiles/tx_katy_freeway.aspx) [Accessed on August 28, 2018].
 Jay Blazek Crossley, It Took 51% More Time to Drive Out Katy Freeway in 2014 Than in 2011, Twenty three more minutes, Houston Tomorrow, May 26, 2015. (Link: http://www.houstontomorrow.org/livability/story/it-took-51-more-time-to-drive-out-katy-freeway-in-2014-than-2011/) [Accessed on August 28, 2018].
 Anthony Downs, Traffic: Why It’s Getting Worse, What Government Can Do, The Brookings Institution, Policy Brief #128, January, 2004. (Link: https://www.brookings.edu/wp-content/uploads/2016/06/pb128.pdf) [Accessed on August 28, 2018].
 Gilles Duranton, M. Turner, The Fundamental Law of Road Congestion: Evidence from US cities,American Economic Review, American Economic Association, vol. 101(6), pages 2616-52, September, 2009.
 Vincent Jarnigon, Bouchons : les bandes d’arrêt d’urgence testées ?, Ouest France, December 13, 2016. (Link: https://www.ouest-france.fr/bretagne/rennes-35000/bouchons-les-bandes-d-arret-d-urgence-testees-4680846] [Accessed on August 28, 2018].
 European Environment Agency,Occupancy Rates, Publications, April 26, 2016. (Link: https://www.eea.europa.eu/publications/ENVISSUENo12/page029.html) [Accessed on August 28, 2018].
 Nathalie Obadia, Corentin De Chatelperron, La voiture reste incontournable dans les agglomérations, LesÉchos.fr, October 8, 2017. (Link: https://www.lesechos.fr/08/10/2017/lesechos.fr/030676421146_la-voiture-reste-incontournable-dans-les-agglomerations.htm) [Accessed on August 28, 2018].
 Grégory Rozieres, Pour réduire les bouchons, il suffirait de 5% de voitures autonomes sur les routes,Le Huffington Post, May 10, 2017. (Link: https://www.huffingtonpost.fr/2017/05/10/pour-reduire-les-bouchons-il-suffirait-de-5-de-voitures-autono_a_22079696/) [Accessed on August 28, 2018].
 Carole Blanchard, Paris : vers la fin du péripgérique ?, BFMTV, November 8, 2017. (Link: https://auto.bfmtv.com/actualite/paris-vers-la-fin-du-peripherique-1298000.html) [Accessed on August 28, 2018].
 ADEME,Étude nationale sur le covoiturage : leviers d’actions, benchmark et exploitation de l’enquête nationale transports et déplacements (ENTD), septembre 2015 [En ligne : https://www.ademe.fr/sites/default/files/assets/documents/etude_nationale_covoiturage_courte_distance-leviers_action_et_benchmark.pdf] [Accessed on August 28, 2018].
 Fondation Concorde, Accompagner la mise en place du télétravail,Synthèse, June 6, 2017. (Link: https://www.fondationconcorde.com/accompagner-la-mise-en-place-du-teletravail/) [Accessed on August 28, 2018].
 Downs notes, however, that this principle does not affect the ability of a road widening to increase the flux of vehicles (number of vehicles that are moving per hour).
 Gaële Lesteven, op. cit.
 Serge-Christophe Kolm, La théorie générale de l’encombrement, Futurible, Paris, 1968.
 Jean-Michel Gradt, Embouteillages en France : une facture estimée à 17 milliards d’euros par an, LesÉchos.fr, October 14, 2014. (Link: https://www.lesechos.fr/14/10/2014/lesechos.fr/0203856715962_embouteillages-en-france—une-facture-estimee-a-17-milliards-d-euros-par-an.htm) (Consulté le 12 août 2018).
 Gaële Lesteven, op. cit.
 Paul Minett, Are Predictable Traffic Jams a ‘Tragedy of the Commons’?,Move Forward, Infrastructure, September 11, 2015. (Link: https://www.move-forward.com/are-predictable-traffic-jams-a-tragedy-of-the-commons/) [Accessed on August 28, 2018].
 Tim Roughgarden, Selfish Routing, Cornell University, May, 2002. (Link: https://theory.stanford.edu/~tim/papers/thesis.pdf) [Accessed on August 28, 2018].
 Shoshana Vasserman, Michal Feldman, Avinatan Hassidim, Implementing the Wisdom of Waze, Proceeding IJCAI’15 Proceedings of the 24thInternational Conference on Artificial Intelligence, 2015.
- Route by Samy Menai from the Noun Project
- Time by Richard de Vos from the Noun Project
- Car by Gregor Cresnar from the Noun Project