Thoughts on Regulatory Failure

Good regulation is hard. Regulatory mistakes are harder. In this post we revisit the regulatory failure surrounding the death of William Ball. In a future post I will outline some ways regulators can further protect the public and themselves against these past mistakes.

Stuart MacKinnon

03 Jul 2025 — 5 min read

Photo by Nik / Unsplash (Not the seatbelt in question).

On Saturday 6 January 2018, 65 year old William Ball (or Billy to his friends) took the front passenger seat of a 1991 Nissan Sentra for an outing that was to be his last.

On State Highway 12 near Turiwiri, Dargaville, driver error caused the vehicle to leave the road into a drainage ditch. William's seatbelt snapped and his head hit the windscreen. He died of his injuries 26 days later in the ICU of Whangarei Hospital. A tragic story of regulatory failure began to unfold.

Police examining the vehicle noticed frayed seatbelts in the car, and due to the failure of William's seatbelt they immediately referred the matter to the New Zealand Transport Agency (NZTA).

On 15 December 2017, less than a month before the crash, the owner of the Nissan Sentra had taken it to Dargaville Diesel Specialists (DDS) for a Warrant of Fitness or 'WoF' inspection – a regular requirement for vehicles on New Zealand roads.

DDS was appointed an 'Inspecting Organisation' by NZTA in 2010 but over the years it had established a long history of poor practice and noncompliance with NZTA vehicle inspection standards.

By the time of the crash, NZTA had made 14 compliance visits to DDS, including the fourth in a series of unannounced visits on 18 December 2017, only days after DDS provided a WoF for the Sentra. In addition, NZTA had received 7 complaints about DDS WoF issuing practices – the latest being an anonymous tip in March 2017 that a WoF had been issued without an inspection taking place. There were questions to be asked.

Independent reviews by Kristy McDonald ONZM KC and MartinJenkins (linked below) provide thorough analysis of events and contributing factors. Rather than span every complex detail, I want to briefly explore a few thoughts that inform my work when assessing regulatory systems and processes.

Regulated parties are not customers

At the time, NZTA had begun to consider its regulated entities, like DDS, its customers. In doing so, it lost sight of its ultimate customer - the public it was there to protect.

Malcolm Sparrow thinks of half the government being involved in the provision of services to citizens, and the other half in the protection of citizens. Of the regulatory and enforcement half he says:

This half of government is different. These agencies use the coercive power of the state, seeking to control behaviors. They deliver obligations, to citizens and businesses, more often than services. Acting in the public interest, they can restrict private behaviors, deprive individuals of their liberty, impose conditions on businesses, put people out of business if necessary, and occasionally kill. Not everyone is happy about the work they do. Not everyone says “thank you.”

Sparrow (2020), p. 4

As a former psychologist, I'd venture that not everyone has the temperament for regulation. It's not a game for hobbyists, or for people who need everyone to be happy. As Sparrow points out, regulators deal with the things that not everyone agrees on, where market forces alone are inadequate. There will be unhappiness, and sometimes you are the person who brings that unhappiness about.

This is not to say that regulators can run roughshod over citizens and businesses. They can't legitimately impose their will without sound evidence of the harms prevented. Nor can they act without being measured when circumstances allow. The dynamic tension of regulation is never resolved – but delivering regulatory obligations is not a service, and regulated parties are not customers.

Pathological empathy?

This confusion between customer service and regulation can be a mask. WoFs bring revenue and additional work to a business in a small community. Depriving a small business of its ability to conduct vehicle inspections could be the same as forcing the business to close its doors.

This kind of dilemma is not limited to WoF provision. There are many kinds of businesses that rely on a regulated service component to function at all. Pulling the pin on someone's livelihood is unpleasant. When I look at the history of DDS's engagements with NZTA, I have to consider that some kind of pathological empathy may have been a factor. Perhaps NZTA's preference to assume a customer service posture was just one outward manifestation of this empathy.

Regulatory theatre

A empathetic customer service posture can easily drift towards regulatory theatre, where the regulator assumes the outward appearance of effectiveness.

Audits are carried out, corrective actions are issued, and licences bestowed. The people are there. The regulatory machinery is there. It just fails to work in concert. The system doesn't provide the protection from harms that an ordinary person would expect.

A regulator puts both the public and the public’s trust in government at risk when it fails to ensure tight controls and clear linkages between: audits; the completion and maintenance of corrective actions; and decisions to relicense a regulated entity.

Astoundingly, even as NZTA's Compliance Intervention Panel was progressing the suspension of DDS's licence (achieved on 27 August 2018), in parallel, another group within NZTA was working with DSS to correct its paperwork issues so it could achieve its next licence renewal.

If a regulated entity has a history of unresolved issues at audit, and they continue to be licensed regardless, the stage is set for a potentially tragic ending. Regulatory systems can tolerate this kind of ambiguity for quite some time. It works until it doesn't.

Normalisation of deviance

Normalised deviance is most commonly associated with the 1986 Space Shuttle Challenger disaster. However, it's a useful lens in regulatory environments that have drifted into regulatory theatre.

Potentially fatal problems with the Space Shuttle solid rocket booster O-rings were observed in 1981, 1984, and 1985. In 1986, management and engineers debated the risk of unprecedented cold weather on the O-rings deep into the night, before a decision was made to go ahead with the fateful launch the following morning.

There's a special kind of confidence that comes from repeatedly and successfully cheating fate.

Like each launch of the Space Shuttle carrying its known defect, it's possible for regulators to begin to feel they are rewriting the principles of common sense each time they are ignored without obvious consequence.

A common feature is for people to maintain a code of silence around the deviance. There are indications in the independent reviews into the DDS regulatory failure that at times front-line regulators were discouraged from initiating rigorous action or raising problematic cases for review and decisions.

What's happened since then?

In the intervening years NZTA has left no stone unturned to address the many complex issues identified in the two 2019 reviews. These reviews initiated extensive change which is documented in the MartinJenkins December 2021 final report. These documents are available on NZTA's website and are linked below.

So why bring this up all these years later?

I've written this because humans never stop being human. For some, our instincts for empathy and our desire to serve can run counter to the requirements of robust regulatory practice.

We are good at telling ourselves comforting stories about why we do things, and at thinking things that happen to others can't happen to us. Even if this is only a brief pause to reflect, it is at least an acknowledgement of William Bell and some of the lessons his tragedy left us.