How feedback saved Virgin Atlantic Airlines $5.37 million dollars, 21.5 million kg in CO2 emissions, and made their pilots happier

Intervention · Employment and Work Performance

Abstract

Airline fuel consumption is costly for both the environment and airlines. Reducing fuel consumption would be a win-win for both. Economists paired with Virgin Atlantic Airways and applied a behavioral approach to see if they could incentivize pilots to use less fuel. As pilots have a degree of discretion over the amount of fuel used, the study looked at how increasing information about fuel consumption, setting personal goals to lower it, and making fuel-saving part of the greater good affected fuel usage. This behavioral intervention was highly effective, resulting in $5.37 million dollars saved and 21.5 million kg less CO2 emissions. 

WANT TO WORK TOGETHER ON A RELATED PROBLEM?

Effective interventions start with a nuanced understanding of how decisions are made. Our mission is to help large organizations be better and do better, using behavioral science.

Learn about what we do

Rating: 5/5 (Significant results; ethical design)

How different behavioural interventions reduced the fuel
savings and efficiency of 335 Virgin Airlines pilots
Conditions
Results
Feedback on fuel usage $5.37 million in overall fuel savings 
Positive Encouragement (e.g., “well done!”) 1.3 minute decrease in fuel consumption per flight 
Altruistic Encouragement (e.g., meet fuel target to donate to Doctors Without Borders.) 6.5% increased job satisfaction

Key Concepts

Incentives: Anything that motivates a person to do something. These incentives can be moral, coercive, or remunerative.

Altruism: Altruism is the practice of making sacrifices for other people’s benefit due to our care about their well-being. Often, these actions offer no direct reciprocal benefit to us, but we feel good doing them. A great example of altruistic behavior is donating to charity.

The Problem

Climate change is the paramount challenge of our time. The existential threat of climate change has made sustainable forms of consumption and production necessary if we want to secure a better future. One way to do this is to cut back on the use of fossil fuels. The current level of fuel consumption has led to an overproduction of CO2, causing an increase in global warming. A lack of preventative action to clamp down on CO2 production will result in ecological devastation.

The major impacts of air travel

Technological advancements and globalization have allowed many to lead lives of great comfort and exploration. Airplanes have proved themselves to be incredibly useful tools for opening up the world to trade, travel, and improved welfare. Unfortunately, they are also heavy fuel consumers. While replacing fuel-powered airplanes with a more sustainable alternative is on the horizon, rapid action must be taken to mitigate climate change’s most pressing effects.

Making flights more eco-friendly

Recent research has shown that pilots have a significant amount of discretion over the amount of fuel consumed by their planes. A study hypothesized that pilots had an optimistic view of their own fuel consumption – thinking that they were already saving ample fuel, they would unintentionally consume more fuel than necessary. A behavioral intervention incentivizing pilots to lower their fuel consumption could be a cheap, immediate, and effective measure to decrease airline carbon footprints.

Design

Economists Gosnell, List, and Metcalfe partnered with Virgin Atlantic Airways to see if a behavioral intervention could be as powerful as predicted. As the group of pilots was highly experienced, unionized, and paid well, the ability to accrue more productivity from them is an impressive feat.

What they measured

The study analyzed three facets of the flying process: fuel load (how accurately the pilots assessed fuel levels prior to the flight), efficient flight (how fuel-efficient the pilot was while in the air), and efficient taxi (how fuel-efficient the pilot was at driving the aircraft after landing).

The 3 experimental conditions

The researchers split their 335 pilots into three groups. The goal of the experimental design was to examine the distinct effects from each incentive on the fuel savings behavior of the pilots. The researchers hypothesized a lack of information was the main reason for overconsumption, and decided every pilot group was to be given feedback on their fuel levels.

  1. The first group of pilots was only given feedback reports on their fuel use performance from the previous month. This, in essence, acted as a control group. 
  2. The second group was given monthly feedback as well, but was also asked to achieve personalized fuel targets that were 25% lower than how they performed the month before.

    • If they were successful in achieving two thirds of their goals, they would receive a simple nudge in the form of a “Well Done!” encouragement at the bottom of their paper.
    • If they failed to achieve their targets, they were nudged by gentle encouragement to fly more efficiently. 
  3. The final group had the same feedback reports and the personalized goals as group two, but the researchers wanted to test to see if the pilots could be further motivated by altruism. They added another dimension: if the pilots could reach their targets, funds would go to a charity of their choice.

Ensuring freedom for pilots

The experiment was designed to be entirely self-directed and at the discretion of the pilots. It used incentive schemes that allowed full flexibility to use whatever means for them to reach their goals, rather than dictating what they should do or inserting another manager.

Results and Application

Millions in savings: financially and ecologically

The results of the intervention were astounding. Overall, accounting for engineering and fuel data considerations, the interventions led to $5.37 million saved and 21.5 million kg reduced CO2 emissions. But which of the interventions contributed most to these results?

What worked best?

Findings showed that significant improvements occurred from simply providing information for pilots about the actual level of their fuel consumption. This supports the hypothesis that pilots do actively try to save on fuel, but aren’t aware of their wasteful baseline behavior. The most effective intervention was the inclusion of personal goals; their implementation drastically decreased fuel consumption in each facet of the flight process.

Charity made for happier pilots

The charitable donation, while still more successful than the control, didn’t seem to make much difference compared to the second treatment. However, when they examined job satisfaction, they found it rose by 6.5% for pilots who engaged in charitable giving. Therefore, even though it may not be the most effective measure for saving fuel, it seems that a prosocial behavioral intervention made the pilots happier. These interventions were low cost and resulted in environmental, corporate, and social benefits.

Related Applications

Industry
Application
Health & Wellness Providing monthly feedback on monitored walking/exercise/drinking habits could be a cost-effective and efficient way to incentivize better levels of physical activity and healthier dietary habits.
Financial Services Banks and fintechs can use monthly goal-setting to incentivize more rigorous and steady saving habits.
Climate & Energy Providing consumers and households with a breakdown of their weekly electricity consumption (based on appliances, lights, etc.) has been shown to drastically increase their awareness of energy usage and lower their overall energy consumption.

Ethics

  • Researchers actively encouraged pilots to monitor their own data and ask what data was collected from them
  • Pilots were consulted on experiment design, allowing for utmost consent
  • Safety was not addressed, despite the general safety concerns related to passenger air travel
Dimension
Verdict
Comments
Welfare

Does the intervention demonstrably improve the lives of those affected by it?
Positive
Pilots used 17% less fuel, reducing carbon emissions on flights. Job satisfaction also went up by 6.5% in the charitable giving condition.
Does the intervention respect the privacy (including the privacy of identity) of those it affects?
Positive
Researchers encouraged pilots to ask what data was being collected on them, and monitor their own data throughout the experiment.  Researchers did not share data with their employers.
Does the intervention have a plan to monitor the safety, effectiveness, and validity of the intervention?
Room for Improvement
Clear effectiveness measurements, but doesn’t address safety.
Autonomy

Does the intervention abide by a reasonable degree of consent?
Positive 
Captains were allowed to choose how much fuel they used and which charities they supported. Consent was not coerced in any meaningful way.
Does the intervention respect the ability of those it affects to make their own decisions?
Positive
Captains were consulted on how to best design the experiment to make sure this autonomy was respected.
Does the intervention increase the number of choices available to those it affects?
Not Applicable
The number of choices stayed the same in most conditions.
Equity

Does the intervention acknowledge the perspectives, interests, and preferences of everyone it affects, including traditionally marginalized groups?
Positive
Captains were consulted before and during the study, to make sure the study was consistent with their preferences. Virgin Atlantic Airways and the captains' union also agreed to the terms of the experiment.
Are the participants diverse?
Room for Improvement
94% of captains were male.
Does the intervention help ensure a just, equitable distribution of welfare?
Room for Improvement
While climate change is, in part, an equity issue, the study does not address it in depth.

Related TDL Content

The Future of Air Travel: Over the course of the COVID-19 pandemic, our relationship with airlines and travel has drastically changed. In this piece, the Decision Lab examines what the future of air travel may be in the wake of the pandemic, and what role behavioral science will play in that transformation.

Precommitment: Part of this intervention’s success was that captains committed themselves to a certain emission goal, leading them to try to reach their emissions target throughout the month. This kind of precommitment strategy has many more applications than just reducing fuel consumption though. In this piece, the Decision Lab walks through what precommitment is, why it works, and where it can be implemented.

Warm Glow Giving: One noteworthy result in this intervention is that, although charitable giving failed to decrease emissions as much as other strategies, it notably increased job satisfaction. This is a textbook case of “warm glow giving:” the positive feeling people get when they think they have “done their part” by donating to, or acting on behalf of, others. In this article, the Decision Lab through what Warm Glow Giving is, when it was discovered, and where it can be used.

Sources

Gosnell, G. K., List, J. A., List & Metcalfe, R. (2016). A NEW APPROACH TO AN AGE-OLD PROBLEM: SOLVING EXTERNALITIES BY INCENTING WORKERS DIRECTLY.NBER  Working Paper Series. http://www.nber.org/papers/w22316.

Notes illustration

Eager to learn about how behavioral science can help your organization?