LSE - Small Logo
LSE - Small Logo

Jens Madsen

June 6th, 2023

Why is change so hard?

0 comments | 20 shares

Estimated reading time: 4 minutes

Jens Madsen

June 6th, 2023

Why is change so hard?

0 comments | 20 shares

Estimated reading time: 4 minutes

People’s habits die hard and many seemingly reasonable change interventions have little or no effect. If those who propose changes sit behind their desks, they will find it hard to understand the lived realities of the people who would be the target of the interventions. Jens Madsen writes that, to avoid failure, proponents must be humble, listen to each other and engage with stakeholders in the target communities.


Jens Madsen will be speaking at the LSE Festival event Why is Change so Hard? Tuesday 13 June 2023 6.30pm to 7.45pm.


 

To generate a better society, people and organisations want to intervene to effect change. Of course, what counts as a ‘better society’ is an intensely political question, which depends on one’s moral stance, what outcomes you hope to see, and how you see other people. While it is extremely interesting to discuss what makes a society good, this is not the point we want to consider here. Rather, we are considering why it can be so hard to effect change on a societal level.

The UK government launched campaigns to increase vaccination rates during the COVID-19 pandemic, The World Bank gave financial support to improve water sustainability in Madhya Pradesh, and The Bank of England have raised interest rates to combat rising inflation. To minimise the impact of misinformation, researchers have developed tools to train people to be better at spotting it. Interventions can be many different things like financial, educational, or political attempts to change people’s behaviours such as getting people to take a vaccine that makes them less vulnerable to a rampant virus. Interventions often target features that are thought to influence relevant behaviours. If you think people care about money, governments may put additional taxes on cigarettes to discourage smoking. However, if you think people value their health, you may put health warnings on cigarette packages. That is, the shape of the intervention depends on what you think about people.

Yet, many seemingly reasonable interventions have little or no effect. People’s habits may lessen the impact of interventions. If they have always smoked and find social pleasure in doing so, they may be willing to risk some health and lose some money to keep doing it. Similarly, if there is a cultural tradition of crop growing, it can be socially precarious to be the first mover – if you choose a new cultivation technique that goes against generational knowledge and your harvest fails, it is not only financially damaging, but also a loss of reputation. Sunk costs, apathy and habits, and loss aversion are powerful motivators to not change, even if the incentives are promising. In this environment, it is no wonder that change can be hard, as the intervention is not only fighting a narrow set of utility calculations, but is fighting apathy, tradition, and fear of loss of face.

In the worst of times, interventions can even backfire. As a set of example of this, The Cobra Effect describes situations where incentives cause people to act opposite to the intended outcome (also known as ‘perverse incentives’). The name stems from an anecdote from British Colonial Rule of India where cobras were a problem. To combat cobras, the British offered money for people who handed in the dead animal. However, people quickly learnt that they could earn a nifty sum of money if they bred cobras to hand in. After a while, the government caught wind of it and stopped the cash-for-cobra scheme. As the snakes were now worthless, the breeders let them loose in nature, which caused the cobra population to increase overall. While the name may stem from an anecdote (and we should therefore be careful in trusting it), there are many reported examples throughout history. In 1973, the US government imposed development restrictions on land with endangered species – so people began to kill the animals for fear of discovery; French Colonial rulers paid for rat tails in Hanoi in 1902, which led to a proliferation of tail-less rats released in the sewers; payment for medical treatment and reimbursing insured patients for treatment but not prevention may encourage medical conditions to be ignored until treatment is required.

There are many reasons why interventions may fail, and systems can seem resistant to change. Here, we consider two: the understanding of behaviours and the complexities of the system. We have considered apathy, habits, and reputational loss, but other preferences can influence behaviour. As an example, consider what makes fishers choose where and what to fish. This may seem trivial, as it is easy to believe it is all about the money. Indeed, bio-economic models often disregard behavioural complexity by assuming that fishers are perfectly informed rational profit-maximisers (Anderson 2015).

When interventions are tested under these assumptions, they are met with two problems: real fishers are neither perfectly informed nor necessarily profit-maximisers. In Wilson’s (1990) words, assuming perfect information “assumes away the problem of finding fish”. Learning where to fish involves trial-and-error, exploration, and generational knowledge. Learning and guessing is a huge part of fishing decision-making and cannot simply be abstracted away. On the second assumption, studies show that fishers may forego profit due to deference to in-group fairness, concerns about conservation, or compliance with social norms. They may rate consistency, sustainability, and neighbourliness as important as income, which may be especially relevant in small-scale fisheries, where social forces can be important drivers. If we misunderstand the motivations for behaviour and people’s capacity to make decisions, it is hardly surprising that interventions play out differently in real life than in model predictions. Understanding this requires collaboration between economists, social and cognitive psychologists, political scientists, and NGOs or other officials who understand the population in question. Quick models based on simplified assumptions from behind a desk risk that interventions fail or backfire.

The second problem is that interventions often play out in complex environments. Complex systems are defined by four core features: interactivity and feedback loops, adaptation, heterogeneity, and the passage of time. When people can interact with each other, their behaviour can cause feedback loops that can change the whole system fundamentally. As an example, consider when stock markets start to panic. If traders start to sell off positions, other traders may look to this to interpret that a crash is coming, so they start selling off their positions. If the media picks this up and reports uncertainty in the stock market, ordinary investors may begin to panic and sell off their assets. All of a sudden, a small initial sell-off from a few prominent investors may cause a mass panic, which results in seismic changes to economies around the world. In this cauldron of activity, people may be more or less risk averse (heterogeneity), interactions may create feedback loops, which may cause people to sell their assets (adapt their behaviour) – this all happens over time, as the system unfolds and changes. In the same way that complex systems can change quickly, they can also be extremely plastic and resistant to change. If the problem is complex, interventions have to consider the ripple effects on the whole system. If we fail to understand the bi-directional relationship between people and the structure of the system, we risk that interventions either fail or have unintended consequences.

Inter-disciplinary work is a key tool to meet this challenge. Economists may understand material conditions that influence people but may fail to appreciate psychological differences that drive behaviour. Psychologists may have a good grasp on socio-cultural preferences but may fail to understand the environmental conditions that set the boundaries for activity. Environmental scientists produce wonderful work to understand physical features of behaviour but may not adequately understand the economic conditions. As an example of inter-disciplinary work, the POSEIDON project brings together economists, ecologists, psychologists, NGOs with experience working in fisheries, and others to create a holistic fisheries model. This has led to diverse work on the use of data to check the model, implementation of new behavioural models, comparison of behavioural models, interviews with Indonesian fishers, testing of policy, and more.

Interventions are all-to-often proposed by academics and officials who are isolated in their way of seeing the world. Being open to inter-disciplinary discussion allows for critical examination of fundamental assumptions. In line with this, field work is critical. If we sit behind our desks of academia or Whitehall, it can be hard to understand the lived realities of the people that would ultimately be the target of the proposed interventions. We must be humble, listen to each other, and engage with stakeholders in the communities to avoid that interventions fail or backfire.

♣♣♣

Notes:

  • This blog post represents the views of its author(s), not the position of LSE Business Review or the London School of Economics.
  • Featured image by Anastasia Vityukova on Unsplash
  • When you leave a comment, you’re agreeing to our Comment Policy.

About the author

Jens Madsen

Jens Madsen is an assistant professor at LSE’s Department of Psychological and Behavioural Science.

Posted In: Career and Success | Economics and Finance | LSE Authors | Management

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.