The end of a policy to support carpooling in Jakarta, Indonesia led to an 87% increase in evening rush-hour traffic, according to a study published this summer in Science.

Vehicles slowed to a 7 mile (11 kilometer) per hour crawl, showing the profound impact of moving away from carpooling restrictions. During the morning rush-hour, a smaller – but still major – 46% increase in travel times was observed. Furthermore, traffic worsened throughout the city, including on smaller parallel roads that had not previously been restricted and also during periods outside of rush-hour.

Jakarta’s carpooling policy had required that cars using the high-occupancy vehicle (HOV) lanes on major roads have a minimum of three passengers during rush-hour periods. This was meant to reduce travel times throughout the city, which is home to more than 30 million people.

But a common practice in the city of hiring “professional passengers” in order to qualify for HOV lane access raised concerns on the benefits of the carpooling policy. According to Rema Hanna, a Professor of South-East Asia Studies at Harvard University and an author on this new study:

"In Indonesia there were professional passengers, called jockeys, who stood right outside the entrance to the HOV areas. So if you were short a passenger you could hire a jockey to be the third passenger, for a little over a dollar. Some people thought that the fact that these jockeys existed made the whole policy moot."

Even in absence of this controversial practice, these types of HOV policies can be controversial, says Hanna:

"The key issue with HOV policies is that they remove existing lanes from use by general traffic … the question is about whether the additional carpooling behavior makes up for the loss in road space."

The high level of controversy is why, in March 2016, the city government announced that it would cancel its carpooling policy in Jakarta within seven days.

Seeing an opportunity to better understand the impact of HOV policies, when hearing this news Hanna and her colleagues jumped onto Google Maps and began collecting data on traffic conditions in Jakarta. According to Hanna:

"Within 48 hours of the policy announcement we were querying data from Google Maps every 10 minutes to check traffic conditions in Jakarta. We included both the main roads affected by the policy change, as well as other roads in the network in order to understand how traffic was changing citywide."

One can access the full research paper here <link>