Nudge theory

Learning from: behavioural economists

How can a combination of psychology and economics, in particular nudge theory, benefit OSH practitioners? 


Conventional economics relies on the assumption that people always make rational decisions, that they have unlimited willpower to convert those decisions into behaviour and will act in their own best interests to maximise their gains. The rational person will save for their retirement, buy only what they need, and will never pay over the odds for a coffee in a disposable cup. If traditional economic theory is extrapolated to other areas of behaviour, people will never overeat, will take exercise and will always follow safety rules.

From the 1950s, economists began to realise that human choices could not always be explained mathematically, so they looked at other disciplines to improve the potential of economics to predict (or at least explain) decisions. In particular, they looked at psychology, sociology and anthropology, and later at neuroscience, to provide a more complete picture of what affects our decision-making. The field of behavioural economics (BE) was born.

On test

How to influence behaviour

The UK’s Behaviour Insights Team’s EAST model doesn’t work in isolation – it needs to be part of a structured process:

  1. Define the outcome. Identify what behaviour is to be influenced, how you will measure it, and what size of change you need over what time period.
  2. Understand the context. Know the people whose behaviour you want to influence and see it from their point of view.
  3. Build your intervention. Is there anything obvious in the hierarchy you need to do? If there are organisational factors that prevent people doing the right thing, or disincentivise the right choices, deal with these before you start considering behavioural insights. Then consider how to make it easy, attractive, social and timely.
  4. Test, learn, adapt. Try it, and measure its effects. Randomised controlled trials are rarely practical in a workplace but if you have two sites you could intervene in one and compare before and after for both.

Some people describe BE as simply rebranding psychological principles. It is true that many of the terms used in the discipline are recognisable as psychological ones. For example, behavioural economists apply the cognitive biases we looked at in a previous article (IOSH Magazine, February 2018, bit.ly/2E2f21a). But they have perhaps been more adventurous than psychologists in testing their theories on large audiences.

A range of BE studies with impressive results is set out in EAST: four simple ways to apply behavioural insights (bit.ly/2HgVLKW), published in 2014 by the UK’s Behavioural Insights Team (BIT), a social purpose company partly owned by the Cabinet Office and also known as the Nudge Unit. The research shows that personalising requests for tax payments makes people more likely to comply and that how you ask individuals to leave money to charity in their will affects not just the likelihood of them making a bequest, but also the size of the gift. But can any of this help OSH professionals to make safer and healthier workplaces?

Down to choices

Just as eliminating a hazard is at the top of our familiar hierarchy of safety and health controls, so too eliminating the “wrong” choice is the top of a hierarchy of behavioural controls (see “Increasing form-filling” box). But just as eliminating hazards is sometimes impractical, eliminating choice is often impossible.

When incentives are removed, if punishment seems unlikely or, if other factors promote the unwanted choice, people will revert to “undesirable” behaviour when no one is watching. In the Behavioural Economics Guide 2017 (bit.ly/2sAqZIP), Cass Sunstein writes: “When people feel that their control is being taken away, they will often rebel, even if exercising control would not result in material benefits or might produce material harms.”

At the bottom of the hierarchy, providing information can be enough to change behaviour. “There are alligators in the pond” might be enough information to stand well back. But if impacts are complicated most people don’t have the time, background knowledge or inclination to process the evidence. The drop in the uptake of the MMR vaccine among parents in the UK because of one scare story in 1998 was a demonstration of this.

Imagine we could persuade managers to crave the experience of walking around the workplace 

Steering people towards our preferred choice – “nudging” – does not rely on providing them with all the evidence, but it must preserve their sense of freedom of choice. We want them to make the “right” choice because they think that is what they want to do, not because they feel compelled. Anurag Vaish, a “behaviour architect” based in Mumbai, India, offers an example that helps to understand the distinction: “A nudge is something subtle, embedded in the environment; it is not awareness dependent. Diet programmes are not nudge. But offering food on smaller plates is a nudge.”

This is a useful example to keep in mind when considering the word incentive. The smaller plate guides the choice, but it is not a financial incentive recognised by traditional economic theory.

Nudge-theory-learning-from-behavioural-economistsThe strength of a nudge is that people form new habits. The dieter learns to eat less at every meal, whatever the plate size; the employee develops the habit of donning safety boots as they arrive at work if the boots are stored in a location people have to pass.

In The Power of Habit (2012) Charles Duhigg describes how the most successful campaigns are those where people crave the desired behaviour. Imagine we could persuade managers to crave the experience of walking around the workplace, that people looked forward to the opportunity to practise an evacuation drill, or that workers felt undressed without PPE. Nudges might not take us that far, but they are a good starting point.

The BIT’s EAST publication suggests four factors to nudge a behaviour:

  • make it easy
  • make it attractive
  • make it social
  • make it timely.

BE can apply to behaviour at any level. Take two occupational safety and health challenges. First, how to cajole workers to evacuate a building quickly in an emergency or a drill. Though they know how dangerous a fire can be, when the alarm bell sounds people stop to finish emails, collect coats and bags, and walk slowly so they can carry their coffees to the assembly area. Second, how to convince senior managers to show their commitment to OSH by having regular, effective safety conversations with workers. Could behaviour economists suggest any ideas to overcome these problems?

Make it easy

Human factors expertise has long been applied to make it easier to do the right thing. Yet we know that errors are still made because of confusing user interfaces, poor process design, and organisational structures that restrict communication. Our two challenges are more complicated than painting a button red – they are about habits and behaviour.

Allison-Reynolds-Consultant-Science-Based-LeadershipThe inverse of making the right thing easier is to make the wrong thing harder. The EAST report describes a study that found deaths from paracetamol poisoning fell by 43% after new legislation required that larger packets provided tablets only in blister packs. The hassle of releasing each pill individually was in many cases enough to discourage the self-harm attempt.

For fire drills, Allison Reynolds, behavioural change coach with consultancy Science Based Leadership, suggests one way to make the wrong thing harder to do. If the problem is people staying to finish work tasks, she suggests locking work machines automatically when the fire alarm sounds – though not for weekly tests perhaps. If people are going to fetch belongings anyway, “make coat and bag storage close to the work area so people can grab them on the way past”, she says.

From Iceland, Lilja Kristjansdottir, a BE enthusiast with a master’s degree in evolutionary psychology, considered the second challenge. She says that if senior managers have to walk through a work area for some other purpose it would be easier to have regular conversations. “If senior managers had regular meetings in the vicinity of frontline workers, this might induce them to have the conversations on the way to the meeting,” she says.

An alternative would be to put the best-tasting coffee in a location that forced managers to walk through the work area. “Plying them with amazing coffee or biscuits might work,” she says.

Lilja-Kristjansdotti-BE-enthusiast-with-a-master’s-degree-in-evolutionary-psychologyTechnology can also help. When paper-based systems are used to record emergency lighting and fire alarm checks, for example, it is not uncommon to find responsible staff telling an auditor: “I did it, I just didn’t write it down”. If tasks are prompted by a tablet-based device which records the event, the default is that the result will also be recorded, making life easier for the responsible staff – and the auditor.

Make it attractive

We don’t need BE to tell us that clearly set out, illustrated procedures, safety notices or warnings are more likely to be followed than badly designed pages of overcrowded text. However, the finding that we comply more when a request is personalised is less obvious, and certainly not applied in what is too often over-generic safety documents. The UK Driver and Vehicle Licensing Agency sent out letters warning that recipients might lose their vehicles if they didn’t pay their vehicle tax. When an image of the vehicle was included in the letter, response rates increased to 49%, from 42% for the pictureless control group.

Similarly, recipients whose names were included in text messages sent out by HM Courts and Tribunals Service were more likely to pay their fines than those receiving a standard text message or those receiving a message that included a reminder of the amount of the fine.

How personalised are the evacuation instructions for your staff? Too many fire action notices don’t even have the fire assembly area indicated in the blank space provided. Since many people now get their safety information from an intranet, it wouldn’t be too difficult to have evacuation information tailored to a location – this is the name of your nearest fire warden, the position of your nearest alarm call point, your exit route, the number of your assembly point and so on.

Make it social

Increasing form-filling

Whether you want people to report near-misses and hazardous situations, to answer a stress or safety culture survey, or to document equipment or system checks, there is plenty of advice on how to improve form-filling completion rates.

Using shading to indicate which parts of a form a user should complete and careful wording of questions are all essential to make the form easy to complete and attractive.

If forms can be completed electronically, auto-populating the information the system knows can increase not just take-up, but the quality of the follow-up. If an employee is already logged into the system, why should they have to enter their name and department? If a mobile device is used to log a fault, the GPS in the device could offer a default location. Dates and times can also be added automatically, yet there are many safety and health forms on which users must add the date more than once and then calculate review dates.

A myth of traditional economics is that our rational decisions are driven by “unbounded selfishness” – that is, we think only of the maximum benefit for ourselves (and perhaps our families). Fortunately, we don’t live in a world where people think only of themselves. Even rats in a maze will stop pressing a lever for food when they discover that it causes suffering to another rat.

Kristjansdottir proposed an imaginative way to use the social nudge for swift building evacuations. In an initial drill, identify the loiterers. Then she suggests: “Ask them to make a commitment to those in need of assistance.”

As well as wanting to help others, our social nature encourages us to conform with what we perceive to be the norm. If others don’t wear PPE or delay moving in an evacuation, our instinct to protect ourselves can be overridden by the need to be the same as everyone else. Of all the BE heuristics, this is one when safety and health might need a paradigm change. Moaning “I can’t get people to wear their PPE” or “No one seems to take safety seriously around here” is telling people that if they don’t wear the equipment they will fit in. Tax collecting bodies in the US and the UK have found an increase in payments when individuals are told that most other people have settled. Similar results have been achieved in encouraging people to give more money to charity, reducing littering and increasing recycling.

So, rather than exaggerating problems, we should celebrate the positives – most people do evacuate immediately and those managers who have had safety conversations have found them useful. Personalising the comparison has been shown to be extra effective – so you might try something along the lines of: “Nine out of ten people in your department arrived at the assembly point within three minutes of the alarm being sounded.”

Make it timely

Timeliness is interpreted in different ways by BE. One finding is that behaviour is easier to change when habits are already disrupted, so an early safety induction will have a greater impact on new staff than providing the same lesson some weeks in. Could you run fire awareness training in the afternoon after a drill in the morning when any difficulties are fresh in people’s minds? If there has been a major accident in the news, could you run an awareness campaign on any relevant behaviour for your own workplace?

A second feature of timeliness is that we are more influenced by immediate costs and benefits than those delivered later. Telling stakeholders that “spending on safety pays for itself” is a platitude to avoid, since the long-term benefit of preventing a potential accident is not as salient as the short-term cost of the redesigned workspace. The immediate benefit of getting on with the job now seems greater than the long-term benefit to health of taking time to find PPE; the benefit of having 15 minutes to work through emails appears to outweigh any important insights that might be gained by spending the same time on a safety conversation. Reynolds recognises the difficulty of making long-term benefits motivational: “Supervisors could pay more positive attention to desirable behaviours, rather than rely on a promised benefit at some point down the line.”

Committing time is also important and the EAST report gives examples of improvements in attendance at medical appointments when people write down the time and doctor’s name themselves (rather than simply receive a card). Tim Houlihan, founder of Behavior Alchemy in the US, suggests an application of this idea to nudging managers into having safety conversations. “The manager’s immediate bosses should ask them to set a time in their diary each week at the same time. A regular commitment becomes part of everyone’s expectations.”

In its EAST report the BIT reminds us: “When applying behavioural insights, we need to recognise that context matters.” The examples given here are to inspire thought, and any ideas should be tested. What works in one context, might not work in another. The website www.miml.co.uk gives a few more ideas that may inspire your own solutions.

BE does assume that someone benign knows what the optimal behaviour is. BE therefore will not help to identify decision-making errors if the wrong behaviours are reinforced. However, when we have evidence of the preferred behaviours – whether at senior level or on the front line – EAST gives us a simple reminder of how to assess any approaches we might try to use.



Bridget Leathley is a freelance health and safety consultant, providing risk management support in facilities, retail and office environments.  She delivers face-to-face safety training including IOSH and bespoke courses, and contributes to e-learning courses through evaluations and design work.  She has been writing for health and safety publications since 1996.  

Add new comment