Skip Ribbon Commands
Skip to main content
Bookmark and Share this page


Remarks before the Coalition of Airline Pilots Association, 2009 National Safety Conference Dallas, Texas
Deborah A. P. Hersman
Coalition of Airline Pilots Association, 2009 National Safety Conference Dallas, Texas

Thank you, John and good morning everyone.  I consider the invitation to address this group an honor.  Pilots like you are the best and last line of defense we have, when it comes to preventing an accident.  Not only do you hold the lives of thousands of passengers in your capable hands every day, but you have volunteered, by your participation in CAPA, your union, and in this conference, to be a catalyst for safety improvements.

I am here today representing my fellow board members, Vice Chairman Chris Hart, Member Robert Sumwalt, and all the men and women of the NTSB.  As many of you know, the NTSB is a small, independent, federal agency.  We are 400 people strong, with about a third of that number involved in aviation.  Our mandates are to investigate transportation accidents, to determine their probable causes, and to issue recommendations to prevent future accidents.  Our independence is crucial.  We call it the way we see it.  We are not regulators, and we simply don’t have the budget to fund the safety improvements that we recommend.  We can effect change only through persuasion—and with the weight of our credibility.  It  is up to others to implement our recommendations.

From my perspective, not only as the 12th Chairman of the NTSB, but as a Board Member for the past five years, I would like to share with you some observations about what our accident investigations have revealed about leadership.  During my time at the Board, I have launched with our investigative teams on 17 major accidents, and I have reviewed the findings and recommendations of over 100 more.  I have come to appreciate that in addition to the necessary skills and knowledge, strong leadership practices are excellent layers of defense in preventing incidents and accidents.

Leadership has a different meaning for each of us; it can encompass a host of definitions.  There are scores of books on the topic, and you heard outstanding presentations yesterday on leadership.

Let me share a quote with you about leadership from someone that probably wasn’t quoted yesterday by the leadership experts.  Rosalynn Carter said, “A leader takes people where they want to go, but a great leader takes people where they don’t necessarily want to go, but ought to be.”   The next presentation is entitled “Good to Great: a line survey of cockpit leadership,” so to help tee up Capt. Swauger’s talk, I’ve modified the quote for CAPA’s conference:  “A good leader takes people where they want to go, a great leader takes people where they need to be.”

I do not pretend to be an expert on leadership, I have a tremendous amount to learn from others.  But I think we can all admit that it is a lot easier to ride the crest of a wave when you are a leader – if I have a group of people who know the direction they want to go, then we can all get there together and hold hands and sing “kum-baya”.  But what about when there is a tough message that needs to be delivered, a situation where your team needs to be held accountable for their performance or you, yourself are part of a situation that lets people who are counting on you down?  Are you a leader that a) knows where you and your team NEEDS to go and b) can you lead your team where they NEED to go?

I am confident that the unions represented here today have committees that address safety and training and procedures.  Leaders don’t wait for someone to tell them to make a change—they take the initiative; they become catalysts for that change.  I believe that all of you are here because you have made the decision to be a catalyst.  Some of you have a passion for safety, some of you are natural leaders who influence people with your vision, your passion, or your professionalism. My challenge to you today is to figure out where you need to lead your team and begin the process of getting them there.

I am pleased to see that Human Factors, Crew Resource Management, Training, and Fatigue are being addressed at this conference.  Although the 50 fatal Colgan accident in Buffalo is still under investigation, I believe that the NTSB’s public hearing in May—not even three months after the accident—served as a catalyst.  It brought these issues to the forefront of the public’s consciousness, including a discussion of the effects of commuting.  The Colgan accident, in my view, is a watershed accident that is motivating the public, the regulator, the industry, and perhaps, even some unions or individuals in this audience to initiate overdue changes.

As you are aware, the public discussion of safety issues following the Safety Board’s Public Hearing in May, the advocacy by victims’ family groups and the concerns of industry and labor groups  motivated Congress to take action.  Two weeks ago, the US House of Representatives passed an airline safety bill, requiring FAA and the airlines to raise the bar on safety through enhanced training, establishing ASAP and FOQA programs and addressing fatigue countermeasures to name a few.  Many of these provisions were the result of NTSB recommendations that have languished for years, such as improved upset training or revising flight and duty time.  While we would like to see a statutory or regulatory “raising of the safety bar” so that all carriers adhere to the same standards, we also recognize that many operators voluntarily adopt more stringent or robust standards – that effort of getting out in front of your peers requires institutional and organizational leadership.  How are you raising the bar on safety within your organization?e

CAPA Accident Data

The Safety Board is the official census keeper for aviation accident statistics.  So therefore we don’t get to tell the good news stories… we discuss what went wrong.  We don’t do it to be judgmental, but because the only way we can learn from incidents and accidents is to discuss where the system broke down…  As I prepared my remarks, I tapped into our database to review the accidents and incidents that the 9 member airlines of CAPA have experienced over the last five years.  I did this, not to put any one organization on the spot, but to give you a snapshot of where your organization might NEED to go with respect to safety improvements.

I will delve into the specifics of two other air carrier accidents that we have investigated to further illustrate the importance of what was presented yesterday, and what will follow later today.  There is a phrase etched in glass at the entrance of our NTSB Training Center in Ashburn, Virginia, that says “From tragedy we draw knowledge to improve the safety of us all.”  I hope leaders like you will find this accident knowledge valuable.

First, the CAPA experience.  I will discuss the NTSB data generically, without citing specific airlines. Over the past five years, the 9 companies that employ CAPA pilots experienced 43 accidents worldwide, plus an additional 32 incidents that the NTSB investigated, for a total of 75 investigations.  Of those, 27 involved a mechanical malfunction that placed the flight crew in a degraded safety position.  In the vast majority of these cases, the pilots’ leadership and skill led to an uneventful landing with no injuries or damages to the aircraft.  Here are some quotes straight out of our files to illustrate this:

  • The airplane experienced a failure of the No. 2 engine while climbing to flight level 330.  The flight crew reported feeling a severe vibration, pulled both throttles back to idle, declared an emergency, and performed an uneventful single-engine landing.
  • A Boeing 747 … was climbing through 15,000 feet … when the No. 1 engine separated from its forward and aft engine mounts, resulting in substantial damage to the airplane.  The flight diverted to an alternate airport where it landed without further incident.
  • A Boeing 737 experienced a rapid decompression while in cruise flight.  The decompression was caused by a 1.5 square foot hole in the crown fuselage near passenger row 20 … the flight crew declared an emergency and the flight landed uneventfully.
  • During cruise flight the crew experienced smoke in the cockpit … followed by the fracture of the inner pane of the first officer’s windshield.  They declared an emergency and diverted the airplane to a local airport where a successful landing was made.

Another notable accident involving a mechanical failure, albeit bird-induced, was, of course, US Airways Flight 1549.  Talk about "Leadership and Safety” and “The Command Perspective!”  Captain Sullenburger and First Officer Skiles provide a sterling example of that.  I talked with some pilots last night that said  First Officer Skiles provided an outstanding presentation last night during your dinner, so I won’t repeat the many positive lessons that the crew of 1549 gave all of us who are involved in this industry.

14 other accidents of your 75 investigations occurred on the ramp.  Unfortunately, these “fender benders” met the definition of substantial damage, so they are counted as accidents.  However, the vast majority of these were caused by the actions, or inactions, of those outside the cockpit.

Another 18 investigations involved inadvertent and sudden encounters with turbulence, in which mostly flight attendants received a broken bone, which meets our definition of a serious injury, and hence, are also classified as an accidents.  These three categories—mechanical, ramp events, and turbulence—account for roughly 80% of  the accidents and incidents that your organizations have experienced.

Now for the uncomfortable part.  It is the remaining 20% that point to areas in which improvement of pilot performance and better leadership in the cockpit  may have prevented the accident or incident.  Here are some more quotes from our files to illustrate this:

  • The captain’s failure to follow company procedures for stabilized approach.
  • The captain’s failure to maintain adequate airspeed during the landing, which resulted in a stall.
  • The flying pilot’s improper aircraft handling after main landing gear touchdown, which resulted in the collapse of the nose landing gear assembly.  Contributing to the accident was the pilot-in-command’s inadequate supervision during the landing. and
  • The flight crew failed to lower the landing gear until notified by the air traffic control tower to go around.

OK – we are human and we all make mistakes – and in my case, I am fortunate enough that most of the time “spell check” catches my errors or I can ask someone to repeat a question if I was distracted.  You, however are on the pointy end of the spear, your error or inattention could result in an incident or an accident that affects hundreds of other people.  If you are lucky you get to share your side of the story, but in the worst situations, you leave behind a 30-minute or 2-hour cockpit voice recorder and an FDR to tell the story to hundreds or thousands of grieving family members and friends of people who were counting on you to get them somewhere safely, not to mention, the pain experienced by your family and friends.  No pilot is immune to a lapse, a mistake or an error, but history suggests that pilots who not only talk the talk, but walk the walk of safety leadership are far less likely to be being cited in an NTSB probable cause statement.  Like Capt Sullenberger and First Officer Skiles, keep making deposits in the bank of aviation safety by doing adhering to SOPs, doing the right thing every day, over and over, because you never know what 30 minutes of flight could define your career.

Leadership and Decision-making – Alaska Airlines Flight 261

The NTSB has investigated many accidents in which overt or even subtle pressures on flight crews have led to poor operational decision-making.  The lessons that aviation leaders can take away is that safety leadership means that safety must come first, even when it’s not the easiest or most convenient choice.
An example of this can be found in our investigation of Alaska Airlines Flight 261, which crashed just off the coast of California 10 years ago, killing all 88 people aboard the MD-83.  The Board determined that the probable cause of the accident was a loss of airplane pitch control that resulted from the catastrophic failure of the horizontal stabilizer jackscrew actuator due to inadequate maintenance.  In our report, the Board cited the good and the not-so-good aspects of how the flight crew handled the pressures of a mechanical malfunction.

First, the good, and this crew was commended in the report for certain decisions.  The flight was en route from Puerto Vallarta, Mexico, to San Francisco, when the stabilizer trim failed.  The crew decided to divert to LAX.  This decision was based on several factors, including more favorable wind conditions for landing at LAX and the captain’s concern about overflying suitable airports.  This is one area where the flight crew exercised leadership by deciding to divert, especially because the airline’s dispatch personnel appeared to have attempted to influence the flight crew to continue to San Francisco.

Comments recorded by the CVR indicated that after the captain stated his intention to divert to LAX, dispatch cautioned that if the flight diverted, it would be “probably an hour to an hour and a half [before the airplane could depart again].  We have a major flow program going right now.”

The captain responded, “I really didn’t want to hear about the flow being the reason you’re recalling us, cause I’m concerned about overflying suitable airports.

Later, the captain commented to a flight attendant, “It just blows me away they think we’re gonna land [and] they’re gonna fix it.  Now they’re worried about the flow, I’m sorry, this airplane’s [not] gonna go anywhere for a while.”

The Safety Board concluded that the flight crew’s decision to divert to LAX was prudent and appropriate.  This is an example of good cockpit leadership … having the courage to make  a decision in the face of potential criticism.  Courage is an important leadership quality.

However, a few minutes later, and after recovery from an initial uncontrolled dive, the captain told air traffic that he wanted to “change my configuration, make sure I can control the jet and I’d like to do that out here over the bay if I may.

The captain then ordered extension of the slats and the flaps.  He did not brief the first officer about what to expect or what to do if these configuration changes resulted in excessive flight control pressures or loss of control of the airplane.  Further, the captain did not specify that the flaps should be extended at a slower-than-normal rate, which would have been a prudent precaution to minimize the possibility of the configuration change causing abrupt airplane movements that could be difficult to control.

Nevertheless, after the slats and flaps were extended, the captain noted that the airplane was “pretty stable.”  But, 9 seconds later, the captain ordered retraction of the slats and flaps, and the airspeed began to increase.  What happened next is depicted in this animation:

[ Slide 11 – Alaska Airline Flight 261 ANIMATION ]

ANIMATION: Available at

While it did not cite this configuration change as contributing to the accident, the Board noted that an airplane with flight control problems should be handled in a slow and methodical manner, and that any configuration that would aid a landing should be maintained if possible. The Board concluded that flight crews dealing with an in-flight control problem should maintain any configuration change that would aid in accomplishing a safe approach and landing, unless that configuration change adversely affects the airplane’s controllability.

These pilots were in a challenging situation, what would you have done?  How would you have communicated with your co-pilot?

Leadership and Flight Crew Monitoring

Another investigation where the Board identified shortcomings in  leadership in the cockpit, occurred on December 18, 2003, when Fed Ex flight 647, a Boeing MD-10, was landing at Memphis International Airport.

Following the hard landing, the right main landing gear collapsed, and there was a post-crash fire.  There were two crewmen and five nonrevenue pilots aboard the airplane.  The first officer and a nonrevenue pilot received minor injuries during the evacuation.

This animation depicts what happened during the landing:

ANIMATION: Available at

The Safety Board determined the probable cause of the crash was the first officer's failure to properly apply crosswind landing techniques to align the airplane with the runway centerline and to properly arrest the airplane’s descent rate before touchdown.  Additionally, and more pertinent to the topic of leadership, the Board determined that the captain's failure to adequately monitor the first officer's performance, and failure to command or initiate corrective action during the final approach and landing, contributed to the accident.  During the accident flight, the captain was serving as both check airman and pilot in command; he was expected to continually monitor the first officer's performance, while at the same time being responsible for the overall safe conduct of the flight.  Since I know many of you are check airmen, I should also note that there have been subsequent accidents where we have identified the fatiguing workload associated with training other pilots.  For anyone in the room that has taught a teenager to drive a car – the mental and physical stress of doing something yourself versus coaching someone else to do it is significant.

[ Slide 15 – Data and percentages on “Monitoring” Accidents ]

In a 1994 Safety Study, the NTSB found that inadequate crew monitoring or challenging was a factor in 31 of 37 crew-caused air carrier accidents that were reviewed.  That is 84%.  The study found that flight crews frequently failed to recognize and effectively draw attention to critical cues that led to the accident sequence.  More recent air carrier accidents, such as a collision with trees on final approach in a Fed Ex 727 in Tallahassee, Florida, and the crash of a Cessna Citation on approach to Pueblo, Colorado, support a our recommendation to the FAA to require pilot training programs be modified to contain modules that teach and emphasize monitoring skills and workload management, and that include opportunities to practice and demonstrate proficiency in these areas.  If your airline is not doing this, you should ask why?  Are you waiting for the FAA to implement this recommendation?  Could you exercise leadership in union committees to effect this change on your own?

Leadership and a Safety Culture

We are hearing a chorus of support for the integration of safety management systems, or SMS, and a realignment of responsibility and accountability for operators as we move toward a more performance-based approach to safety.

The NTSB has issued several recommendations since 2007 addressing the importance of SMS programs.  For example, in 2007 we asked the FAA to require that all Part 121 operators establish SMS programs.

This recommendation was issued following our investigation into the 2004 crash of Pinnacle Airlines flight 3701 near Jefferson City, Missouri.  While there were numerous safety issues identified in our investigation, because your conference is focused on leadership in the cockpit, I will focus on the operational areas identified in the report. The NTSB determined the probable causes of this accident, in part, were the pilots’ unprofessional behavior, deviation from standard operating procedures, and poor airmanship, which resulted in an in-flight emergency from which they were unable to recover, in part because of the pilots’ inadequate training.

The Pinnacle accident was my first launch, it served as my training accident, and then about 9 months later I chaired a 3-day public hearing on the accident.  If you read the CVR transcript, the accident will serve as a stark example to you about what happens when there is a lack of discipline and leadership on the flight deck.  Early in the flight the FDR documented that the crew were exceeding normal parameters, before the got the 15,000 feet, the crew switched seats, they were dispatched on the repositioning flight to fly at flight level 330, but they requested clearance to take the aircraft to flight level 410, they laughed and joked about the angle of attack while monitoring the aircraft climbing 500 ft per minute, they stalled the aircraft, got into a PIO/APC, flamed out both engines, didn’t correctly follow restart procedures, didn’t declare the true nature of the emergency – at FL360 they declared a single engine out – it wasn’t until they were below 10k feet at which time they had overflown several suitable airports that they declared double engine out, they didn’t make it to Jefferson City and crashed in a residential neighborhood.  Sadly this crew lost their lives in this accident.   What enabled the crew to believe that it was o.kay to fly the way they did?  Were they an anomaly or the norm as far as their knowledge, skills and professionalism?  What recommendations might prevent a future event?  The Safety Board made numerous recommendations following the accident, on the operational side, recommendations addressed upset training, Safety Management Systems, and implementing a FOQA program – a lack of a safety culture was at the heart of some of these issues.

The Safety Board recently provided comments to the FAA’s Advance Notice of Proposed Rulemaking (ANPRM) on SMS, which was published in July.  The NTSB is encouraged that the FAA is considering requirements for SMS implementation, not only for Part 121 operators, but for commuter and on-demand operators, training providers, maintenance repair stations, and product manufacturers.  However, the ANPRM does not specifically address corporate operations under Part 91 or fractional ownership operations under Part 91 subpart K.  The NTSB encouraged FAA to include these operations in SMS rulemaking to allow them to benefit from the proactive management of safety by implementing SMS programs.

The FAA specifically requested comments on the role of voluntarily submitted safety data through programs such as the Aviation Safety Action Program (ASAP) and how these programs would fit within an SMS framework.  The NTSB considers programs such as ASAP and the Flight Operational Quality Assurance (FOQA) program to be fundamental tools for operators to realize the safety assurance component of SMS programs.  The NTSB has a longstanding interest in programs such as ASAP and FOQA, and has issued safety recommendations encouraging the adoption of these programs.

Yesterday, you heard from Sidney Dekker on the importance of “Just Culture”.  Does your organization have a just culture?  I won’t ask if your organization has a safety culture, because everyone has a safety culture, it may be a good one or a bad one…What kind of safety culture does your organization have?  Is it something you are proud of?  You have an ownership stake in that culture – is it where you want to be, where you need to be?  If not, how can you take it up a notch?  Do you know what needs to be done?  Can you do it?

I firmly believe an SMS can help a good organization become better, but to be a good organization, that wants to be better, requires an underlying positive safety culture.  An organization lacking a robust safety culture can check the boxes of an SMS, but likely won’t enjoy the benefits that come to those that embrace improvements and incorporate them into the fiber of the organization.


Thank you for your attention and thank you for your continued commitment to safety leadership.  Thank you for the roles you play as check airmen, as analysts in safety programs, as incident or accident investigators, as safety advocates within CAPA, your company, on the Hill, and for leading at the very local level – in the cockpit on each and every leg you fly.

No matter what your role is, I’d ask you to consider this when you leave this afternoon:  where does your organization need to go?  What role are you going to play to take them there?