Blowing Up The Status Quo

How might the aviation accident investigation process be improved? That is the question the National Transportation Safety Board (NTSB) asked for public comments in an August 24 announcement posted in the Federal Register. Comments must be submitted not later than October 14; that is not much time, given the importance of the subject.

But, then, from the NTSB’s 18-page notice, it is obvious that substantive changes to the current process are not envisioned. The NTSB proposes sundry technical edits to existing regulations. Left intact is the direct affront to the NTSB’s charter as an independent investigative body, the so-called “party” system. Under this existing arrangement, airlines, manufacturers, air traffic controllers, pilots and mechanics unions and, not least, the regulator — the Federal Aviation Administration (FAA) — are accorded an official role in the investigation. A representative for the killed or injured passengers is not included. However, those involved in the deaths or injuries of passengers are official members of the investigation.

The official definition of a party is “a person, government agency, or association whose employees … or products were involved in the accident or incident and who can provide suitable qualified technical personnel to assist in the investigation (49 Code of Federal Regulations, paragraph 831.111(a)(1)).

This arrangement has come about partly because the NTSB is a small agency, without enough investigators to do the job. The agency has about the same number of aviation investigators as the equivalent strength of a Marine Corps rifle company. That is 130 Marines which, in NTSB “uniform”, are charged with airline accident and incident investigations, which includes air charter operations, plus investigating the same for the nation’s private, general aviation, fleet of aircraft. All told, the Safety Board has to investigate about 2,500 aviation accidents and incidents yearly with an extremely limited staff.

The party system supposedly has evolved to allow the Safety Board to leverage resources to meet its broad and complex mission.

This may be so, but the parties are hardly impartial. Airplane manufacturers have an interest in minimizing any design flaws in their aircraft that either caused or contributed to the accident. Airlines have in interest in downplaying pilot training or maintenance deficiencies. Pilots’ unions have an interested in demonstrating that the pilots in the accident were not at fault and, indeed, were let down by inadequacies in airplane design or airline practices. The FAA has a vested interest in obfuscating any shortages in its oversight of the airline or shortcomings in its certified approval of the airplane’s design.

Richard Kessler, who lost his wife in an airline crash, takes a dim view of the party system. “As one who has experienced accident investigations first-hand … it is my view that the current process is shot through with conflicts of interest.”

“We can do better,” he maintained.

While parties are not involved in analysis and the determination of probable cause, their intimate involvements is spelled out in the NTSB’s Aviation Investigation Manual. A few extracts from this seminal document serve to highlight the parties’ deep involvement in the warp and woof of the supposedly “independent” investigative process:

“The IIC [NTSB Investigator in Charge] will … meet daily with the party coordinators ….These meetings should be used as a way of determining the parties level of satisfaction with the investigations and their ability to cooperate with each other and with the Safety Board … Guidance and input should be provided to them, as needed, to prevent potential problems from escalating.” The parties’ level of satisfaction is a sure give away that the NTSB will accommodate party wishes.

“The [Safety] Board’s administrative practice regarding the conduct of public hearings is to ensure that before opening the accident file to the public, all parties to the investigation are given the opportunity to record in writing whether or not they want a public hearing to be conducted on the accident.”

“[The] IIC should also e-mail a ‘warning memo’ to the parties notifying them that the docket will likely be opened on the date selected by … the Office of Public Inquiries.”

“The IIC/hearing officer should inform all parties of the details of the hearing and the prehearing conference well before both are held … The IIC/hearing officer should allow sufficient time (at least 2 weeks) for the parties to receive and review the items critical to the hearing before the prehearing conference. These include exhibits, issue list, areas of questioning, and relevant exhibits for each witness.” The public is not accorded this generous courtesy.

“The witnesses may be from the parties to the investigation or can be suggested by one or more of the parties.” If ever there was a set up for friendly testimony, this section of the manual is a dead give away.

“The prehearing conference allows the parties to review plans for the hearing and provides them with a final opportunity to make suggestions on its conduct.” To “make suggestions” and the party’s “self interest” are evident.

“The IIC will write each party … informing him or her of the opportunity to present party submissions. It should be explained that the submissions should contain their proposed conclusions, probable cause, and recommendations relating to the accident … The party submissions will be considered during the development of the final report.” Note the term “will be” taken into account as the final NTSB report is written.


Parties (foreground) participate in an NTSB accident investigation. Why?

In 2005, the RAND corporation submitted a report to the NTSB on its investigative process, a report requested by the NTSB, and observed: “The party process presents inherent conflicts of interest for entities that are both parties in an investigation and ‘parties defendant’ in related litigation.”

This incestuous arrangement can be dispensed with. The party system ought to be expunged entirely. If the NTSB does not have enough staff to handle the workload and the often arcane technical issues involved, there are two remedies, both of which should be adopted:

First, the NTSB should not be involved in the great majority of general aviation accidents. Most of these involve pilot error and/or poor judgment (e.g., taking off with inadequate fuel or persisting in a flight where the weather is worsening). This one step would reduce the NTSB’s workload from 2,500 investigations to about 500 yearly.

Second, it should be noted that the NTSB has frequently gone past the parties, hiring small, expert companies, consultants and university professors to fulfill needs for specific technology and knowledge beyond the ken of the NTSB’s professional staff.

The parties no more would be involved in negotiating the issues to be investigated or not to be explored, as is the case at present. The parties now submit a statement of probable cause or suggested recommendations. This astonishing practice would end.

Unfortunately, in its August 12 announcement, the NTSB takes the opposite course, actually liberalizing the role of parties, allowing them to “release information within party organizations as needed to implement prevention [or] remedial action” as a result of disclosure during an accident investigation. Note, this activity is proposed even before the NTSB has concluded its investigation. To be sure, it is useful to accelerate the pace of remedial action, but the real problem is the glacial pace of FAA correctives when in receipt of the final investigation and its many recommendations.

The NTSB proposal abets the status quo, when that arrangement itself is deeply suspect. The status quo needs to be blown up. The often devious and manipulative parties can be dispensed with entirely. If the present parties have a comment, they can submit a letter to the Safety Board, just like any other member of the public. And if the NTSB wants them to testify, a subpoena will serve.

NTSB independence and impartiality as an aviation accident investigation body finally would be assured.