US FAA chief Steve Dickson has defended the administration’s decision to wait for empirical evidence to order the grounding of the Boeing 737 Max, rather than follow other authorities’ precautionary approach.

The FAA grounded the type on 13 March, three days after the loss of an Ethiopian Airlines aircraft – the second fatal Max accident – having held out against a wave of suspensions by other regulators, including the European Union Aviation Safety Agency.

US regulators had originally insisted that there was no basis on which to order a grounding of the Max, despite concerns over similarities between the Ethiopian accident and that involving a Lion Air jet five months earlier.

Speaking during a briefing in London on 6 February, Dickson acknowledged that regulatory alignment would have been preferable.

But he also defended the FAA’s decision to wait for data to establish a common thread, the behaviour of the Manoeuvring Characteristics Augmentation System, between the two Max accidents.

“If you ground an airplane arbitrarily – if you’re making any kind of safety decision arbitrarily – you really don’t know when you’ve got to a point where the situation has been improved,” he says.

“These two accidents had different factors associated with them – two airlines, two groups of pilots – so they weren’t the same scenario.

“They did have a common thread of MCAS. But having the data from which to make those decisions certainly focuses your effort.”

Dickson says has a “big focus” on data at the FAA and believes the industry needs to “raise the bar” and ensure interested parties have the “same kind of availability of data”.

“I don’t know on what basis EASA made their [grounding] decision,” he says. But he believes that the FAA and Canadian regulators took data-based decisions.

“I do know the agency was looking to identify a common thread, and it took getting the data to be able to make that decision. It was not available for a couple of days,” says Dickson, adding that reinforcement of data provision around the world would contribute to moving from “forensic” to “pro-active” analysis.