How Incentives Drive All Decision-Making
The right incentives can help people work together, force organisations to drive positive change, and even help build entire new economies.
From the small scale to the huge scale — incentives matter. They can help one individual save more money and at the same time create a $700bn cryptocurrency ecosystem.
Breaking down the structure of incentives we’ll look at:
- How incentives can help people act in their own best interests
- How incentives can help people work together
- How incentives can help organisations function
- The risks associated with having the wrong incentives
How Incentives Can Help People Act In Their Own Best Interests
Behavioural economists understand how irrational people really are, and how they often need to be ‘nudged’ in the right direction to help them make the best decision for themselves. Using timely incentives is one of these nudges.
So when Walmart wanted to help users of its prepaid card save money, rather than simply offering a regular savings account, they introduced a ‘prize savings’ feature meaning that if the user kept some of their balance in a virtual “vault”, they would be eligible to win a cash prize in a monthly drawing — up to $1000. Every dollar in the MoneyCard Vault would equal an entry in that month’s drawing.
The idea of a lottery-style win incentivised users dreaming of a new car to save small amounts of their paycheck every month.
Another example of how users can be incentivised to save is the use of ‘mental accounting’ to help users compartmentalise their spending and visualise their savings goals. Users can assign separate ‘pots’ of money for spending and saving, with names and photos attached to the pots assigned to savings goals, like saving up for the deposit on a home.
If you can create a separate account called house deposit with a picture of your ideal house on it, you are much less likely to take money out of the account when you’re buying another round at the pub.
How Incentives Can Help People Work Together
In any negotiation, it is crucial to understand the different parties’ incentives. Knowing what the other side ultimately wants enables you to structure a deal that results in a positive-sum transaction for both parties, rather than zero-sum.
If you’re the founder of a company looking for funding, make sure you’ve thought about your potential investors’ incentives before signing any deal. If you want to build a company slowly that will endure over a century and never be sold or listed in the public markets, don’t try to raise venture capital.
Plenty of deals can be made in business where different incentives can result in a great result for both parties — trading one person’s money for the attention of another’s community, or one person’s technical expertise for another’s sales knowledge.
How Incentives Can Help Organisations Function
The right incentives are necessary to help all the employees of a company drive towards the same goal.
How Fedex resolved its shipping delays is a great example of how incentives matter in big organisations.
The heart and soul of the Fedex system is having all their planes come to one place in the middle of the night for workers to shift all the packages from one plane to another. If there are delays, the whole operation can’t deliver its core product — quick delivery.
Despite this, there were consistently delays, and management had tried everything to get employees to work more effectively, appealing to their morals and threatening them, but nothing worked.
Finally, somebody got the idea to pay all these people by the shift rather than by the hour, and allow workers to go home as soon as the work is finished. The problems disappeared immediately.
When incentivising employees, make sure your incentives are aligned with the company’s core KPIs and values. This will prevent employees optimising their effort towards promotions and pay rises rather than corporate success.
This is most evident in Sales teams. Hubspot initially targeted the number of deals closed when giving commission to their salespeople. However, they quickly found that although the number of deals increased, customers would often churn once the commission was paid out. To address this, management put in place a minimum customer contract period before the commission was paid. This ensured not only a high number of new customers but high quality as well.
In big organisations aligning incentives can be even more challenging. If you are a leader trying to enact change in a big company you need to ensure all employees are incentivised to cooperate in achieving that goal. If you want the company to innovate, but individual departments have been tasked solely with cutting costs, then you can’t expect the heads of those departments to pay extra to fund a project that will result in huge cost savings in 3–5 years. If their bonus is paid this year, they won’t want to endanger this year’s targets. Instead, you could create an innovation budget for each team, or judge a leader’s performance based on how well they’ve cooperated with the internal innovation team.
On a much larger scale, new ways of using incentives can even create new economies at scale. The reason why cryptocurrencies have scaled so successfully where other open source decentralised systems have failed is due to the novel use of ‘token economics’. Chris Dixon states that “well-designed token networks include an efficient mechanism to incentivize network participants to overcome the bootstrap problem that bedevils traditional network development”
By granting a larger proportion of tokens to those who do the early development work for a cryptocurrency, you are financially rewarding those people who get the project off the ground and make the application useful to others.
The Risks Associated With Having The Wrong Incentives
In recent years we’ve seen misaligned incentives have global ramifications. While Facebook’s initial goal was to make the world more open and connected, their incentives forced them into behaviour that resulted in exactly the opposite.
The need for growing profits as a public company with a business model based on advertising meant that Facebook was incentivised to drive user attention, not deliver human connection.
Humans are naturally drawn to ideas that confirm their own beliefs and are more likely to click on emotive headlines, so Facebook began to optimise a user’s newsfeed for emotive articles that confirmed a users’ existing political and cultural beliefs. As a result, we’ve ended up with a populace more divided than ever.
Misaligned incentives are also why we need government regulation to prevent negative environmental externalities that are often the result of corporates seeking profit. If corporates aren’t incentivised to look after the environment by the threat of fines and criminal proceedings, they will instead follow their natural incentive — profit for shareholders.
As we increasingly rely on machines for decision-making, we need to think deeply about the incentives and potential second-order consequences that are programmed into these machines. The classic example of this is Nick Bostrom’s thought experiment in Superintelligence about a paperclip maximiser. A machine is designed to accumulate as many paperclips as possible ends up destroying humanity without any evil intention.
While this particular example is unlikely, we are already starting to see the dangers present when machines are simply given data to learn from and asked to make decisions based on correlations. We all know that correlation does not always equal causation, and the resulting machine-led decisions could be harmful.
One example of this is a 2015 study demonstrating that a machine-learning technique used to predict which hospital patients would develop pneumonia complications worked well in most situations. But it made one serious error: it instructed doctors to send patients with asthma home even though such people are in a high-risk category. Because the hospital automatically sent patients with asthma to intensive care, these people were rarely on the ‘required further care’ records on which the system was trained. As a result, the highest risk patients would be sent home.
And there are other less life-threatening examples: Google’s sentiment analyser labelled being a Jew or homosexual as negative, while SAS’s algorithm to rank teachers in the US was making ranking decisions with no description as to why teachers received a particular rank.
Evidently, incentives matter. Whether you are helping people help themselves, work with others better or driving a company towards a goal, a clear understanding of the incentives present in any situation, and how to use them most effectively for your advantage, will help in every aspect of your life. Along with using incentives for positive change, we must always be aware of the risks present when incentives are unintentionally misaligned with our goals, and think critically whether this could be the case in any new initiatives.
