There’s little doubt that New York City’s congestion pricing scheme, which charges vehicles to enter parts of the city, will benefit pedestrians who have to compete with smaller 4,000-pound hunks of metal on wheels blowing through intersections. In fact, it’s also good for the people behind the wheels of the animals. According to a new working paper published by the National Bureau of Economic Researchtraffic backups also decreased in New York’s outer boroughs and suburbs, resulting in shorter travel times even for those outside the congestion zones.

The study used Google Maps traffic trends and travel data to determine average travel times before and after New York City implemented its congestion measures in Manhattan’s central business district. As expected, cars run faster in metro areas with less road traffic. The study found that traffic speeds have increased about 15% since congestion pricing took effect, with greater gains during normal rush hour compared to the period before the toll. Cars that choose to pay the $9 toll and drive through the congestion zone save about three minutes per trip and a collective 83,000 hours per week, according to the researchers.

Contrary to some concerns, however, traffic doesn’t just pile up outside the congestion zone as people try to get around it. However, trips are also getting shorter on the streets. According to the study, traffic speeds outside the business district increased by about 8%, with neighborhoods closer to the congestion zone seeing larger increases. Drivers who avoid the toll zone save a collective 461,000 hours per week in traffic. The study found that the average trip is about eight seconds faster than in the pre-congestion zone, but there are more than 100 times more of those trips than there are trips through Manhattan, so the aggregate savings are higher.

While the biggest effects of congestion pricing occur closer to the zone, the research found that there are at least some broad effects in the toll zone. They found that car trips on Long Island saw speeds increase by 2.3% and that there was even some improvement on highways throughout the tri-state area. Basically, traffic speeds have increased everywhere. The study also found “no evidence of offsetting slowdowns on different road types,” suggesting that the policy “reduced the overall volume of traffic rather than simply shifting congestion.”

That may not mean everything to some of the congestion zone’s biggest detractors, who hate it on principle more than bad policy. President Donald Trump has suggested that his administration kill the congestion zone experiment—even by himself The Department of Justice said in a leaked memo without much legal charge for doing so. You’d think there would be little reason now to have evidence of broad success without significant failures, but when has that stopped this administration from pursuing worse outcomes for everyone?