What Slows Urban Mobility in India — and How to Fix It

Wharton professor Gilles Duranton explains why congestion pricing won't improve mobility in urban India.

Urban economic development indicators such as higher income levels and higher motorization rates are generally associated with better overall mobility, chiefly because people can spend their time more productively than idling away in traffic snarls. Among the tools urban planners typically use to increase mobility is congestion pricing, such as New York City’s recent decision to impose tolls on vehicles entering its busiest areas.

However, such mechanisms will not work in urban India, where planners must focus more on so-called “uncongested mobility” than on reducing actual congestion caused by vehicular traffic. Uncongested mobility essentially refers to the speeds vehicles can reach after navigating everyday obstructions, such as pedestrians, hawkers, stray dogs and cattle. That is the central finding of a research paper titled “Mobility and Congestion in Urban India.”

“One thing that surprised us the most is that on average, across urban India, there is not actually that much congestion,” said the paper’s co-author, Wharton real estate professor Gilles Duranton. “Traffic flows are slow, but they are slow all the time.” The paper’s other co-authors are Prottoy A. Akbar from the University of Pittsburgh; Victor Couture of the University of California, Berkeley; Ejaz Ghani, lead economist at the World Bank; and Adam Storeygard of Tufts University.

Notably, they found that urbanization and rapid growth don’t have to lead to gridlock. Indeed, economic development can bring better infrastructure and also improve mobility. And while India does have “a big mobility problem,” in nearly all cases, the authors said, it is not congestion but other hurdles — bad roads, bad design and potholes — that slow people down. As such, vehicle speeds within the cities showed little variation at different hours of the day — rush hour or not.

Fast and Slow Cities

Differences in urban mobility, or traffic speeds, also vary widely across large cities. Indeed, a factor of nearly two separates the fastest and slowest cities. Chandigarh, a planned city that is the joint capital of the northwestern Punjab and Haryana states, is “a particularly fast city,” while Kolkata (formerly Calcutta), capital of West Bengal state in the east, is “a particularly slow city.”

Another interesting finding was that mobility is better in cities with a more grid-like network such as in Chandigarh. “We were the first to provide the result that the shape of the road network determines how fast you can move,” said Duranton. That’s what the authors call “uncongested mobility.”

“On average, across urban India, there is not actually that much congestion.”

The study also noted that travel is generally slow in Indian cities, even outside peak hours. “Even the central part of Chicago — considered a slow city and one of the most congested locations in the U.S. — is generally faster than one of the fastest Indian cities, Chandigarh,” the paper stated.

Moreover, the researchers found that traffic is slower in denser or more populated Indian cities. Such cities display “a hill-shaped relationship between city per capita income and mobility,” which means that even as per capital income rises, mobility tends to rise up to a point, after which it declines. That means “a city’s mobility is related to characteristics of its road network” and not income levels, the paper said.

When Congestion Pricing Doesn’t Work

The study noticed that wide variations in mobility across Indian cities were primarily due to uncongested mobility, not congestion. As such, congestion pricing would have little impact. That’s why a 10% improvement in uncongested mobility, for instance, could bring more gains than implementing congestion pricing in urban India. Here’s another case: “Due to slow uncongested mobility, a very poor city like Varanasi (Benares) is slower than average at all times, even at night in the absence of traffic,” the paper said.

Duranton said it was not entirely clear to him why urban India has slow mobility in the middle of the day or later in the day. “In most parts of the world when traffic is slow, it is because there are lots of vehicles,” he said. “In India, you have many users on the roads — not just vehicles, but those that use them for parking, for selling and consuming stuff, children playing, and animals, which is a nontrivial issue.”

The paper’s findings challenge conventional wisdom that traffic congestion is the main reason why some cities are slow and some are fast. In fact, Kolkata showed up as the least congested of the four big metros (the others are Mumbai, New Delhi and Chennai), but it was also the slowest because of low uncongested mobility.

“This distinction has important policy implications, because uncongested speed cannot be improved by congestion pricing, ride-sharing promotion or restriction, or other policies often proposed to combat congestion,” the study said. Added Duranton: “Kolkata is extremely slow even at 3 a.m. in the morning. There is no way you can go more than 25 km an hour even when there is nobody on the roads.”

However, congestion-alleviating policies do work in some cases. Duranton said they could be useful in some of the most congested cities, such as Bangalore and Mumbai, but they are unlikely to make much of a difference in cities where traffic is chronically slow, such as Varanasi. “Uncongested speed in the slowest cities cannot be improved by ride-sharing, congestion pricing or other restrictions, and it may be more beneficial to invest in infrastructure,” he said. While the study did not isolate the exact causes of slow traffic, it pointed to poor infrastructure as the likely culprit. “Cities with more primary roads and those laid out in a more regular grid are faster,” he added.

Improving Urban Mobility

Duranton examined some ways in which urban mobility could be improved in India. “First, there is a case to be made for improving roads in urban India — improving mobility irrespective of congestion,” he said. “The low-hanging fruits are probably better management of roads, better traffic management and making sure people are not blocking intersections.”

Subways and rail will bring higher carrying capacity, Duranton said. However, subways are expensive to build. Ambitious subway projects (called the ‘Metro’) are underway in New Delhi and Mumbai. But there is no guarantee that they will decongest the roads by any meaningful degree. “The problem is, as soon as you free up space on the roads, there are 20 cars waiting to use it,” he said. Increased transit options such as subways and rail don’t necessarily reduce congestion because “if you improve travel conditions, more people will travel,” he noted.

“Kolkata is extremely slow even at 3 a.m. in the morning.”

Building new roads is a costly proposition, so a far cheaper option might be fixing potholes and “preventing people from parking everywhere,” Duranton added. However, the economic gains from those measures are probably limited, he said. For example, traffic in New Delhi moves at 15 km an hour, but reduced congestion will not increase speeds to more than 18 km or 20 km an hour, as evidenced by speeds “even in the middle of the night,” he added. “It is not a huge gain.” By contrast, traffic speeds in the center of Chicago average 30 km an hour.

The researchers believe their new approach to studying traffic conditions may be useful throughout the developing world, where there has been a lack of comprehensive data about urban transportation. Urban planners would find useful pointers in the study. “Often we find large shopping malls or industrial parks coming up with scant regard to the traffic density their locations could take, creating congestion,” said Duranton. “Other options such as dedicated lanes for freight traffic, bike trails, pedestrian sidewalks and well-planned traffic signals could also help.”

However, some of those solutions also bring challenges. “Better traffic management is a no-brainer,” Duranton said.  However, reserving roadways for different uses have pros and cons, he noted. Having dedicated bus lanes and bike lanes, for example, has drawbacks as well because of enforcement issues.

“People are likely to use them as temporary parking spots, so it needs to be enforced, and that needs to be designed,” Duranton said. “Bus lanes could increase the carrying capacity of the roads but if you keep one lane free for one bus to go every 30 minutes, I am not going to be convinced that that is a solution. In order to be used efficiently, you need 2,000 travelers per hour. Unless you are able to do that, your bus lane is not potentially efficient.”

The same challenge applies to dedicated bike lanes. “If you kill one traffic lane so you can have 20 bicycles an hour, I am not sure it is a good idea,” Duranton said. “Bike lanes should be on smaller streets; it will also be safer for them.”

“As soon as you free up space on the roads, there are 20 cars waiting to use it.”

A big challenge in India and across the developing world in Latin America and Asia is the increased use of motorcycles, which Chinese manufacturers have made more affordable. But he said the bikes also cause pollution and accidents. For instance, in Bogota, their number has shot up “from zero to a third or half the number of all vehicles” over the past decade, he noted.

Significance of the Study

The study’s findings are based on data culled from more than 22 million trips in 154 large Indian cities over two months in 2016. The authors say their research provides the first systematic investigation of urban mobility across cities in a developing country. In a first, the study’s researchers used a novel data infrastructure to obtain “systematic views” of the actual mobility of people, which Google Maps provided at a low cost.

The study is important for three reasons. “First, there is an extreme paucity of useful knowledge about urban transportation, especially in developing countries,” the authors said. As a first building block towards a more serious knowledge base on urban transportation, some “stylized facts” are needed, they noted.

For instance, there is a need to know how slow is travel in developing cities beyond anecdotal evidence offered by travelers. Policymakers also need a clear understanding of the differences between cities, between different parts of the same city and across times of day within the same city. They hope their results, methodology and data sources can help guide policy and future research on urban transportation in developing countries.

The study’s findings challenge the popular view that urbanization and economic development lead to ever larger cities and increased rates of motorization, which eventually lead to complete gridlock. To be sure, it found evidence of congestion in the largest Indian cities and a strong association between congestion and household access to motorized vehicles. “However, economic development also brings about better travel infrastructure, which facilitates uncongested mobility,” it stated.

Another takeaway for policymakers from the study is that better mobility is associated with a more regular grid network and more primary roads. That clarity helps because “providing an assessment of the determinants of mobility to guide policy is fundamental,” especially when urban transportation in developing countries is prioritized for massive investments, the paper said.

“In India, you have many users on the roads — not just vehicles.”

Indeed, transportation is the largest sector of lending by the World Bank and represents more than 20% of its net commitments as of 2016, the authors pointed out. “The lack of sufficient urban land for building roadways is widely perceived to be the chief cause behind slow mobility and urban congestion, and much of the investments try to address that,” they stated.

Future Research

The study raises three challenges, the authors said. One, there is a need for a new methodology, or a new approach, to measure various forms of mobility from trip information, and to decompose them into uncongested mobility and delays caused by congestion.

Second, they highlight the unavailability of a comprehensive source of data about urban transportation in Indian cities. For their study, they designed a trove of trips and sampled each trip at different times on different days, or what they call “counterfactual trips.” Those were not actual trips, but they mimic some characteristics of trips taken by households in other countries. As it happens, they found that their city mobility indices showed little variance from those they mimicked.

Another challenge was in consistently defining and measuring the cities they tracked for trips. Their solution was to rely on a wide variety of sources including the Census of India, the free OpenStreetMap and satellite imagery. However, a shortcoming of their methodology was its inability to capture the “richness” in the data they would have gained through a survey, Duranton said.