Is it really about culture or is that just what you want to be true?

Forge Works, self-described Global leaders in safety thinking and practice, posted a short clip of what I assume was a longer discussion video, which for some reason piqued my curiosity.

David Provan, a respected expert in his field, stated that commercial airlines have such an impressive safety record that, statistically, flying is safer than walking.

The clip concludes with the thought that this is due to the culture present in what he terms ‘high reliability’ organisations.

I, being of a busy mind and a limited filter, chose to comment on the video to opine that I didn’t think this was necessarily a result of culture, but potentially has more to do with a combination of the level of regulation and the threat to the survival of an organisation in the event of a failure.

As is often the case when the concept of ‘safety culture’ is questioned, those who have built businesses that centre around ensuring organisations buy into the idea that ‘safety leadership’ and ‘culture change’ are the holy grail of risk management suddenly appear like the shopkeeper in ‘Mr Ben’ and challenge anything that might oppose their business model.

One such representative responded to my position, stating that they had personally observed that culture was indeed the key and that the reason other transport operations didn’t meet the same safety standards was due to the lack of this level of leadership/culture in these organisations, rather than the difference in the level of regulation, oversight or reputational harm should an incident occur.

This seemed to be a very narrow and simplistic approach to understanding why aviantion has been able to become reliably safe so I thought it would be worth digging into the safety record of commercial aviation in the hope of discovering the key to their success and whether, it is genuinely due to some sort of mystiocal force that was might call ‘culture’ or, as I suspect, a complex blend of commercial, reputational and internationally applied powerful regulatory oversight.

In true chicken and egg tradition, which comes first: a culture of being safe or the continuously evolving context in which commercial airlines operate? Are they safe because of the industry’s safety culture, driven by internal industry leadership, or is that just another example of simplifying a hypothesis to meet a marketing talking point?

I think a good place to start is with the concepts highlighted in the short video and to investigate whether they hold up to scrutiny. Before I do that, I think I need to make something clear. I’m aware that the published video is a tiny part of a larger discussion, and I don’t have any nuance beyond the clip. I’m not trying to win a debate here, just highlighting the importance of critical evaluation and showing that, with a little digging, perfectly reasonable concepts can be misleading. Bear that in mind before reading what follows.

First point raised: There are many competing commercial aircraft manufacturers that all work to the same high standards.

It makes sense that the more manufacturers there are, the greater the difficulty in maintaining and monitoring standards.

In reality, there are 26 manufacturers of commercial aircraft worldwide, and the market is dominated by only 2. Boeing and Airbus. It is estimated that between 95% and 99% of all commercial airline passengers travel on planes built by these 2 manufacturers.

I’m not aware of any other form of transport with anything like this level of market dominance, but aside from that, it does appear that this makes it feasibly possible to effectively regulate, at least, the design and testing of aircraft.

The next point raised concerns the size and complexity of the supply chains and whether they reliably meet safety standards. As far as I can tell, about 1500 companies across 30 countries are involved. While it is indeed an impressive undertaking, it is dwarfed by motor vehicles and other comparable transport industries. It is also telling that it is an extremely specialised industry that cannot easily be scaled, as it relies on highly skilled staff. This is highlighted by the backlog of about 17,000 jets, equivalent to about 12 years of production.

The next concept is that these aircraft are flown by many different pilots and yet few errors are made.

Becoming a commercial airline pilot is quite the undertaking. It takes about 2 years of experience before attaining a commercial licence, and if you want to fly something like an A380, you are realistically looking at 15 years of airline experience.

There are fewer than 400,000 commercial pilots worldwide, of whom about half are rated for commercial jets.

So what we have is a small number of highly experienced and competent people who, let us remember, never work alone or outside of their rated aircraft abilities.

On a related point, the idea that there are many different airports, each with its own unique environment to navigate. Well, that depends on what you understand by ‘many’, but the number of commercial airports worldwide is fewer than 3,000. About 16% of all air traffic passes through the top 20 locations.

For me, the key point is that not every pilot lands at every airport. It’s estimated that most pilots land at 10 to 20 locations annually.

The next problem to solve is how to enable pilots to fly in a wide range of conditions with minimal impact on flight success. The weather is probably the one thing that the industry does not control. Even King Canute couldn’t stop the tides, and no matter how big the industry, we are all at the mercy of the natural world. What has been developed, though, is excellent weather information, including radar. Aircraft generally fly around anything that looks dodgy rather than through it. There are limits to what aircraft are rated to handle, and often landing somewhere else is the preferred option rather than risk attempting something beyond the aircraft’s or pilot’s design capabilities.

The final point raised is that multiple languages are used. While local dialects are employed in domestic settings, English must be used for international communication. Not only that, but the airline industry has a specialised, coded form of English known as Aviation English, designed to further minimise the risk of miscommunication.

I’d like to emphasise that I don’t dispute that all of this has developed gradually, nor do I want to downplay the significance of what the video is attempting to explain. The fact that such a level of cooperation has emerged is impressive, and the results speak for themselves, but can what has emerged be attributed to culture?

A good place to start is always defining what culture means.

Thanks to ForgeWorks for responding to my comment and clarifying what culture is in this context. “learned patterns of behaviour relevant to safety that are consistent across a group of people’. So protocol, procedures and operational discipline are also behavioural patterns.”

This is a nice place to start. It’s great when culture is defined. At least then we have a common place to start from.

I think this is the cause of most misunderstandings, not only on social media but also within industries and organisations. We are all hearing different things, and nearly always based on our own personal agenda, whether we know it or not.

Culture is a big thing in safety. Many consultants will talk about safety culture, leadership culture and organisational culture. Culture change is promoted as a solution to a multitude of issues. This has driven many organisations to develop and market culture change programs. From “hearts and minds’ to the McKinsey 7-S, to apps and surveys that promise to give leaders something to measure culture against. More often than not, it is based on some type of numerical data points.

The less well-advertised aspect is that culture is an emergent phenomenon, according to researchers in the field. Change to culture arrives, often unexpectedly and rarely predictably, in response to both internal and external factors. To say that an entire industry has a particular type of culture is essentially meaningless. The reality that even within a single organisation, there will be multiple cultures.

The other quick point I’d like to raise here is a personal bugbear of mine: the use of the term ‘holistic’ in general, especially in relation to culture. It is a word that has, in recent years, been used to describe everything, including moisturiser, digital apps and a certain jade egg made to appeal to a specific demographic of the more money than sense variety. It has become a catchword for considering the whole, not just the individual parts. The problem with this use is that it assumes that we can know everything, something I’d argue we can only do in hindsight. It seems to have developed into a cliche rather than adding anything useful to understanding.

Let’s take a step back from culture for a while and try to identify the clues that establish the current operating context for commercial airlines.

It is easy to assume that commercial aviation is a complex operation. There are, after all, many moving parts, and, at first glance, it seems anything but linear.

I find this a lot in my own consulting work. There is an assumption that everything is complex and therefore requires a unique approach to risk management. This, combined with the belief that culture is a driver rather than an output of the operating environment.

The simple fact is, to understand the operating context, you have to know where you are now. At this moment. To do this, you also need to understand how you got here before trying to work out where you can reach from your current position and, importantly, what is, for now at least, a bridge too far.

The history of commercial air travel began in earnest after the First World War. The first scheduled passenger service began in 1914 between Tampa and St Petersburg in the US.

At this time, the industry was essentially unregulated. This led to many crashes and limited information sharing between operators and manufacturers.

The Air Commerce Act of 1926 was the first major legislation. Driven by the US Department of Commerce and early aviation advocates like Orville Wright. It gave the US federal government authority to certify aircraft and pilots and investigate accidents. It was the foundation of what we have now, but the ability to enforce the standards was very limited.

In the 1930’s, two incidents stand out as catalysts for the first legislation with teeth. In 1931, a crash killed 8 people, including an American football coach. This was the first example of widespread public outrage, putting the industry in the spotlight. This was followed in 1935 with the politically explosive death of a sitting US senator in a crash over New Mexico.

These incidents forced Congress to act, and the resulting Civil Aeronautics Act of 1938 created the Civil Aviation Authority (CAA).

After the Second World War, air travel exploded. Faster, higher-flying aircraft operated in crowded airspace with virtually no coordination. What could possibly go wrong?

In 1956, two aircraft collided over the Grand Canyon, killing 128 people. Unsurprising really, when you discover that, at the time, there was no requirement for aircraft to fly under air traffic control.

President Eisenhower’s administration and congressional pressure followed once again amid public outrage over the collision. This led to the creation of the Federal Aviation Agency (FAA, later to become the Federal Aviation Administration), centralising control of US airspace for the first time.

The 1960’s brought Jet aircraft into the mix, with faster speeds, higher altitudes, and new failure modes. The FAA, like almost all regulators, was ill-equipped to keep pace with technological changes.

1960 also saw a mid-air collision over New York, which resulted in the deaths of 134 people. This piled the pressure on the FAA’s newly centralised airspace management mandate.

1974 saw 346 killed due to a cargo door design flaw. The door latch had a known defect that the manufacturer had been warned about. This incident led to the first widespread use of Airworthiness Directives (ADs) as mandatory, rather than advisory, fixes. This was arguably the first change influenced by collective international pressure and litigation, rather than any single government.

In 1977, 583 people were killed when two Boeing 747s collided on the runway. The investigation revealed that crew communication failures, miscommunication with air traffic control and issues with authority to make decisions were evident.

This led directly to the development of Crew Resource Management (CRM) training. Initially introduced by NASA in 1979, and later mandated by the FAA for all US airline crews. It revolutionised the understanding of cockpit dynamics.

1979 saw 273 killed when an engine on a DC10 separated on takeoff due to improper maintenance. The FAA grounded all DC-10s, a dramatic regulatory intervention. This incident strengthened maintenance oversight rules and the FAA’s authority to ground entire fleets.

A Boeing 737’s fuselage ripped open mid-flight due to metal fatigue in 1988. This was found to be caused by the high number of pressurisation cycles on a busy short-haul route. The investigation found that the aircraft was old and had far exceeded its expected operational limits. This, yet again (are we seeing a trend yet?) Caused public outcry and drove the US Congress to implement the Ageing Aircraft Safety Act of 1991. This legislation had the FAA’s backing and mandated structural inspections of older aircraft.

The 1990’s saw further regulatory impact, including the implementation of hydraulic system redundancy standards following the investigation into the 1994 crash, where 132 people were killed due to a rudder hardover, a hydraulic anomaly causing loss of control.

Another incident in 1996, where 110 people were killed after improperly stored oxygen generators ignited in the cargo hold. The crash exposed dangerous gaps in oversight of low-cost carriers. It was a key driver of the FAA Reauthorization Act of 1996 and led to the adoption of mandatory fire suppression systems in cargo holds.

Again in 1996, another flight resulted in 230 deaths when an aviation fuel tank exploded during a flight off Long Island in the US. The investigation attributed the incident to arcing in the wiring near the fuel tanks. This led to changes in fuel tank flammability standards and electrical wiring regulations worldwide.

The events of September 11, 2001, fundamentally reoriented aviation regulation away from safety toward security. Again, showing how the external operating environment has a major impact on the way the commercial airline industry operates. The Aviation and Transportation Security Act (2001) was passed within 77 days of the terrorist attack on the World Trade Centre, creating the Transportation Security Administration (TSA). The US Bush administration drove this through with unanimous congressional support.

In 2009, 50 people were killed in a regional airline crash caused by pilot fatigue and inadequate training. The families of the victims became an extraordinarily effective lobbying force, pushing directly for legislation to prevent the recurrence of this type of incident. The result was the Airline Safety and Federal Aviation Administration Extension Act of 2010, which significantly raised the minimum flight-hour requirement for first officers (from 250 to 1,500 hours) and strengthened fatigue regulations. This is considered one of the clearest examples of victim advocacy directly shaping aviation law. Again, the driver of change was not within the industry, but as a result of external pressure.

Bringing us up to date, the well-publicised Boeing 737 MAX crashes (Lion Air and Ethiopian Airlines, 346 killed in total) exposed what has been described as a regulatory failure. Boeing had developed the MCAS automated stabilisation system, which, in certain circumstances, could repeatedly force the nose down. The FAA had delegated significant certification authority back to Boeing itself, a process called Organisation Designation Authorisation (ODA). Essentially, the FAA chose to trust the manufacturer’s culture due to a lack of expertise to properly scrutinise the system.

The outcome was the Aircraft Certification, Safety, and Accountability Act (2020), the most significant aviation safety legislation for many years. Key provisions limited Boeing’s self-certification authority, required independent safety experts to review critical systems, and mandated direct communication between the FAA and Boeing engineers (bypassing management). Driver for this change were varied and included congressional committees, international regulators (EASA in Europe acted independently of the FAA for the first time in decades, grounding the MAX separately), and the families of victims.

What I find interesting about all of these examples is that the changes were reactive, not proactive, and were almost entirely driven by entities outside of the industry. Families, political pressure, and public outcry have shaped the way commercial airlines operate far more than any internal culture.

That is not to say there aren’t huge numbers of conscientious people working to improve safety within the industry; it’s just that these people are often responding to a changing operating context. Culture is the response to external constraints, not the driver of them.

The modern battle cry of ‘we need to change the culture’ and the many organisations promising leadership training and culture road maps are often guilty of getting companies to put huge amounts of effort into pulling the cart while the horse ambles along at the back.

What can we learn from this? Well, here are 3 things worth exploring:

1. To understand how your industry operates, you have to understand where you are now and how you got here.

2. That culture is an output, not an input, and it is an emergent, iterative process. It is also very unstable and dangerous to rely on ‘culture’ as a control in risk management.

3. Adding constraints helps shape change and can substantially improve the operating environment.

Commercial airlines used to operate in what I would describe, using the Cynefin framework, as a complex, possibly even chaotic, environment. Essentially, it was a state where understanding how each part interacted with others was impossible, and operators were in a continuous state of flux. Looking at the industry today, it seems to be a far more ordered and complex environment. In this setting, there is a link between cause and effect. Although the relationship isn’t straightforward, it requires highly experienced and knowledgeable individuals to manage it effectively, but it remains a stable system. The safety record confirms this stability.

So, what do we think? Can the high safety standards in commercial aviation be attributed to industry culture?

Well, that depends on what you mean by culture. If you mean the collective response of those directly and indirectly affected by the many well-publicised failures, then yes. If it refers to the fact that manufacturers, operators, and air traffic control (among others) are fulfilling their legal responsibilities and that the enforcement of these rules is consistent almost universally across the globe, due to both the strength of the regulator and the potential cost of failure to an organisation. I’ll concede that it’s true.

If you think culture at an individual organisation level or even at an industry level would have manifested in the same outputs in the same timescale without the constraints that have been applied to it through public pressure, investor fears, political influence and regulatory oversight, I’d be genuinely interested in how you came to this conclusion and would love to research some examples that demonstrate this position.

Just a brief note to conclude. This isn’t intended as a criticism of Forge Works or David Provan, both of whom are more knowledgeable, influential, and indeed successful in their field than I will ever be.

It would be unfair to draw any firm conclusions from a short LinkedIn clip. What I aim to achieve by writing this is to broaden the discussion from culture change towards more practical solutions to the issues workplaces face today. This includes not only how ‘work is done’ but also how ‘industries can use constraints’ to enhance outcomes.

To steal from Dave Snowden, who stole from Frozen 2.

We need to “Do the next right thing,” learn from what emerges from that intervention, and repeat the process.