Aviation safety vs the “prosecutorial imperative”. Indiscriminate prosecutions erode safety culture

Safety Information Protection - and learning from accidents

A good way to start the discussion might be by reference to the Montreal Convention of 1999. My guess is that the people in this room know better than anyone what things were like before that treaty came into force. Under the old Warsaw/Hague regime, airlines had strict liability for mishaps, but the victims of an accident were entitled to no more than some absurdly low recovery amount, depending on the jurisdiction in which they were eligible to sue. Even in the United States, where airlines were compelled by regulators to increase the damages available through the treaty, the maximum recovery was $75,000 per passenger.

The only way claimants could break those limits was to prove in court that the carrier had been guilty of "willful misconduct" - a gross negligence, reckless endangerment test that engendered many years of costly litigation that was excruciating for claimants and defendants alike.By the mid-'90s, the airline industry had had enough.

In 1996, through inter-carrier agreements brokered by IATA and the Air Transport Association of America - today's A4A - the airlines waived the liability limits of Warsaw/Hague. The Montreal Convention of 1999 effectively ratified that waiver. Today, strict liability is still the centerpiece of the regime, but unless the airline can prove that the accident was not due to its own negligence - in other words, prove that it took all available measures to prevent the accident -- claimants are entitled to recover all provable economic damages. The net result is that, as long as a claim falls under MC99, there is no longer any reason to spend years fighting in court over whether the airline was guilty of "willful misconduct."

Even in the very rare case where the airline successfully asserts the non-negligence defense, claimants are entitled to 113,100 special drawing rights, or about US$160,000 at current conversion rates.The Montreal Convention made the recovery process more humane, to be sure. But it had an even more important benefit. I know that the plaintiffs' bar prides itself on using the evidentiary tools available in a trial to tease out important facts that might otherwise have gone undiscovered.

But we lawyers have a professional responsibility to construe those facts in ways most beneficial to our clients. Fact-finding thus can take a back seat to advocacy. Thanks to the Montreal Convention of 1999, the litigation-driven incentive to construe the facts in ways most beneficial to one side or the other have largely gone away.

After all, every accident, however regrettable, represents an opportunity to make flying safer, as long as we can find out what actually happened. We can say, thanks to these important developments in civil litigation over the past 20 years, that we now have a much better chance of exploiting that opportunity fully.

You should know that since 1996, when the airlines first waived the limits of liability under Warsaw, the fatal accident rate in commercial aviation has steadily declined. In fact, 2014 was the safest year we have ever had. Jet-hull loss per million sectors flown in 2014 was 0.23, the lowest on record. Whether or not you believe that the elimination of "willful misconduct" trials was a factor in that steady decline in accidents, at least we know that the reduction in such litigation had no adverse safety consequences.

The Prosecutorial Imperative vs. "Just Culture"

But there's another worrisome impediment to learning from occasional mistakes. It is what I will call the prosecutorial imperative.

In too many jurisdictions, the instinct is to treat every accident as a possible crime. There is immediate tension between the technical accident investigators who simply want to find out what happened in the interest of making sure it doesn't happen again, and the criminal investigators who want to determine whether the accident was attributable to culpable conduct and if so to punish that conduct. I don't have to tell you what happens when the gendarmes put yellow tape around the scene of an accident and start quizzing witnesses.

Those closest to the event and with the most valuable information hire lawyers and are warned that anything they say may be used against them. Getting the facts becomes much harder.

Even more worrying is that the prosecutorial imperative can compromise in a fundamental way the over-arching safety ethic that has been so successfully embedded in the DNA of the aviation industry. I'm talking about the "just culture" approach that is widely treated within the industry as a sine qua non to optimal safety performance. The idea dates back to the 1970s, when the first voluntary reporting systems were established.

It is a simple concept: Companies and their employees are encouraged to report voluntarily any defect, any anomaly, any departure from the norm, anything that might compromise the safety of flight. The information and its source are held in strict confidence. And no punishment follows - either of the employee or of the company. No protection is accorded to criminal activity, of course, but short of that, the information remains sacrosanct.

The great thing about the just culture approach is not merely that it produces timely information that can save lives, but that the information is widely shared among those who can benefit from it. There are searchable online databases containing massive amounts of vitally important safety-related information. With today's sophisticated analysis and artificial intelligence, it is possible to predict incipient dangerous conditions and remedy them well in advance of an actual system failure.

An Emerging Consensus: deficiencies are best addressed by a "just culture" approach

The value of just culture has been widely acknowledged by regulatory authorities. The FAA last June issued a new "compliance philosophy" (FAA Order 8000.373, June 26, 2015) that places new emphasis on non-punitive means of rectifying deviations from regulatory requirements when disclosed. Noting that some deviations arise from factors like flawed procedures, simple mistakes, lack of understanding, or diminished skills, the FAA believes that such deficiencies "can most effectively be corrected through root cause analysis and training, education or other appropriate improvements to procedures or training programs for regulated entities...." In other words, not through the imposition of penalties. The objective, quite clearly, is to encourage more voluntary reporting in the interest of ensuring that the safety management systems required of all airlines are working optimally.

Just last month Australia's Civil Aviation Safety Authority issued a new statement of regulatory philosophy that even more explicitly embraced the just culture approach. The agency wrote: "CASA embraces, and encourages the development throughout the aviation community of, a 'just culture,' in which people are not punished for actions, omissions, or decisions taken by them that are commensurate with their experience, qualifications and training."

Earlier this month, the European Commission convened a meeting in Brussels to introduce a "European Corporate Just Culture Declaration." The Declaration said: "It is acknowledged that, in an operational aviation industry environment, individuals, despite their training, expertise, experience, abilities and good will, may be faced with situations where the limits of human performance combined with unwanted and unpredictable systemic influences may lead to an undesirable outcome."

There's a bumper sticker that makes the same point in fewer words. It can be paraphrased as "Stuff happens."

The declaration then continues: "Analysis of reported occurrences by organisations should focus on system performance and contributing factors first and not on apportioning blame and/or focus on individual responsibilities...."

Very clearly, there is an emerging consensus -- among regulatory agencies and the industry -- that encouraging voluntary disclosure of safety information is in everyone's interest, and that the best way to do so is to apply non-punitive remedies to deficiencies that are voluntarily disclosed.

ICAO and Protection at the Global Level

Despite this consensus, however, we have seen too many cases in recent years in which judges, prosecutors, and plaintiffs' attorneys have sought access to this vitally important safety information. In a growing number of instances, they have succeeded. If that trend were to continue, you can be assured that the essential flow of safety information would simply dry up.

This danger is increasingly understood and it's now an issue that's being tackled globally, most importantly at ICAO.

Five years ago, an ICAO High-level Safety Conference recommended the development of new guidance - what ICAO calls "Standards and Recommended Practices" or "SARPs" - to be included in a new annex to the Chicago Convention devoted to safety management. The annexes to the Convention, as you probably know, are where the high-level principles enunciated in the treaty are turned into more specific and granular guidance. They aren't self- executing; they have to be implemented through national laws and regulations in order to be effective, but that's generally what happens. It happens because the quality of a government's aviation safety oversight is measured by the extent to which it has implemented ICAO's SARPs and other guidance.

The new SARPs envisioned five years ago were to spell out government responsibilities for the protection of safety information. The protection of information derived from accident investigations was already addressed to some extent in the accident investigation annex -- Annex 13. The new SARPs were intended to reinforce those protections and explicitly cover information reported via the safety management systems that are now a mandatory ingredient in airline operations - including, of course, the voluntary reporting I've been talking about. This new guidance will be included in the new safety management annex -- Annex 19. And lest there be any doubt, the protection contemplated is protection from prosecutors, judges, and yes, even trial lawyers.

Some of the most important provisions can be found listed under new "Principles of protection" proposed for Annex 19. The first principle is that "States shall ensure that safety data or safety information is not used for: a) disciplinary, civil, administrative and criminal proceedings against employees operational personnel or organizations; b) disclosure to the public; or c) any purposes other than maintaining or improving safety; unless a principle of exception applies."

The "principles of exception" are what you would expect - cases in which the conduct in question clearly crosses the line from an honest mistake into the area of reckless endangerment, gross negligence, willful misconduct, or whatever you want to call it -- conduct that would always be subject to prosecution under applicable national laws.

But the overarching idea, simply put, is that penalizing honest mistakes merely impedes the flow of valuable safety information and thereby actually increases the risk profile of the aviation sector.

ICAO is moving towards a basis for a standard global approach by end-2016

The new provisions were circulated to governments for a final review last July in something ICAO calls a "state letter." Any further comments from the governments were due a week ago, by October 15. The next step will be a review by ICAO's Air Navigation Commission with the intention of presenting the language to the ICAO Council - ICAO's governing body -- for final approval next March. Nobody expects to hear any dissent. The new provisions will then become effective in November of next year.

There is still an open question as to when the new provisions will become applicable to governments - 2018 or 2020 are the options being discussed. As I indicated earlier, nothing in an ICAO annex is self-executing; to be effective and enforceable, the guidance has to be translated into national law by governments. My guess is that a great many governments won't wait for the new language to become effective but will start their legislative processes working even sooner.

All of this is good news for the airlines, of course, but it is even better news for their customers - including you and me. Aviation is already the safest mode of transportation, and by a wide measure. But air traffic is predicted to double over the course of the next 20 years.

That means that we have an obligation to do all we can to make the remarkable safety management systems we rely upon today even better. The changes in law that I've discussed will be an essential element in that improvement.