All of us cheat a little but Volkswagen have cheated a lot.
As the FT have reported the scale of Volkswagen’s dishonesty appears exceptional. Indications are that the ‘defeat devices’ were introduced from 2008 and then adapted to work in different models. 11 million vehicles may have been involved and the cars have been emitting up to 40 times the nitrogen oxide pollutants than is allowed in the US. Given the scale and sustained effort it seems unlikely to be the work of a small rogue team.
So what do we know from Behavioural Economics that can provide some insight into how this unprecedented dishonest behaviour occurred? As the owner of two Volkswagen cars – both diesel – I declare more than a academic interest in understanding why! Here are some likely factors:
1. Failing ‘mandatory’ US NOx standards was not an option
Pass and fail targets increase both performance and cheating. As highlighted in Freakonomics US targets for schools have driven a significant minority of teachers to adjust the scores of students to reach minimum standards and so retain funding.
Volkswagen’s access to the US market was dependent on cars passing America’s stringent NOx standards. It appears that this could not be achieved without a loss of performance and fuel economy that would have made them uncompetitive. In Europe the gap between test and real world emissions appears to have grown since mandatory targets for fleet cars was introduced.
2. Test results below the on-road reality is the industry norm.
Consider the excuse of Lance Armstrong that everyone was cheating or the differences in behaviour you see between professional golfers and footballers. In football it has become acceptable and arguably necessary to exaggerate the impact of tackles.
It appears that emissions from cars built by almost every manufacturer have been higher out on the road than when tested in labs and the gap has been growing. (See the chart below sourced from article in Quartz ). Governments have also been less than zealous on pollution legislation. Volkswagen behaviour differs in how it cheated adopting an illegal approach.
3. Company culture allowed it to happen.
We look to senior management and colleagues to see what is acceptable and right – this supports the group or social norm. Volkswagen’s culture remarkably has allowed the plan to be created, the device to be developed and for implementation to continue. Consider the number of stakeholders engaged, meetings, tests and the data produced to implement this.
4. The loss from stopping was much too big to discontinue.
Small acts of cheating can lead to big ones. Successful results from initial bad behaviour can prime you for more. Tests have shown that prisoners when reminded they were criminals are more likely to respond dishonestly.
In revealing their cheating Volkswagen would have embraced the cost of compensation, damage to brand reputation and loss of market share. Finding performance improvements that removed the need for the devices would have been a more attractive option for all involved.
5. They weren’t stealing money from someone.
Very few of us will contemplate stealing the office ‘petty cash’ – that is removing money from a box looked after by someone. But many more us will be more comfortable borrowing a few post-it notes or pens that over time could add up to tidy sum. Detached from the cash and a victim we will see the cheating as less significant. In the case of Volkswagen the impact of cheating was more pollution. This is bad for our health, but something economists and governments themselves struggle to cost.
5. The risk and costs of discovery were under-estimated.
We all have a tendency to focus on short-term benefits at the expense of those that are less immediate. We are also inclined to an optimism bias that will encourage us to down-play risk.
This is why effective risk management is so important in organisations and in the case of Volkswagen were presumably excluded from the decision. Had someone put a number to the risk anywhere close to the 6.5 billion euros now set aside to help cover the cost of the scandal people would have thought again.
So some understanding but no excuses…
Exploring these biases can help us explain why those in Volkswagen felt the need to cheat and were given the opportunity. By diagnosing what has motivated and enabled behaviour we can identify solutions to reduce the risk of it being repeated.
As a Volkswagen owner I am equally clear that these biases and influences offer no excuses.