How do we help people (or ourselves) to save money
It is really important to have financial savings - a Covid-hit 2020 highlights this. But how do we encourage people, especially those without a habit of saving, to do so?
Almost immediately, as with many other problems, we think of information. Financial literacy education. If only people know how important it is to save, and how much savings accumulate with compound interest, and how quickly savings can run out with one bad hit, everyone will save more.
Except that education has consistently proven ineffective in many fields, including financial literacy. This meta-analysis - a study that looks at all the research conducted on financial literacy - finds that only about 3 to 4% of people act on their finances immediately after going for a financial literacy course. As time passes, this figure steadily drops towards zero.
This shouldn't be surprising. Do we really think people don't save because they don't know savings are necessary? Do smokers smoke because they don't know cigarettes lead to a higher incidence of lung cancer? Do drink drivers ever think they shouldn't drive because they might injure someone or themselves?
Most of the time, we know what the right thing to do is. We just find it difficult to do it
So how do we get people to save more?
Ariely, Robalino et al aimed to find out through this study.
A little context - the researchers worked with the very poor in a slum in Kenya called Kiberia. Even though they had very low income, it was important for the Kiberians to save because they are particularly vulnerable to shocks in life - an illness, a bad harvest, repairs. When these shocks happen, they often have no choice but to borrow. In Kiberia, borrowing is costly - interest is up to 10% per week. If the residents had some savings, they won't need to pay such a heavy price for borrowing.
An additional point - if even those who have a lot less can be convinced to save, then surely findings apply (at least in part) for everyone else with more income.
The researchers worked with a local bank to set up a Savings Plan for their experiment - a bank account that was extra easy to set-up and deposit savings, but much more difficult to make a withdrawal (added layers of bureaucracy).
They then introduced several different conditions (in different combinations) to test which worked the best.
These conditions include:
Baseline condition: Like a normal bank account. No regular communications from the bank
Control condition: The bank sends reminder texts to the participant to make savings
Top up conditions:
- 10% additional top-up at the end of the week for those who meet their savings target at the end of the week
- 20% additional top-up at the end of the week for those who meet their savings target at the end of the week
Top-up conditions with loss aversion: unlike the previous condition, the top-up occurs at the start of the week, and is removed if the participant does not meet a savings target. Thus instead of "gaining" a top-up, in these conditions, participants stand to "lose" a top-up that is already in their savings account.
- 10% top-up at the start of the week, remove for those who not meet their savings target at the end of the week
- 20% top-up at the start of the week, remove for those who not meet their savings target at the end of the week
Instead of a reminder text from the bank, the reminder is sent by the kids of the participants
You can see a picture of the coin above. From the front, the coin carries an image of making savings, while the back carries an image of all the savings that have accumulated. The coin also displays numbers representing each week - these can be marked off when someone meets their savings target. You can see the coin in the picture above.
So which condition do you think worked best? No reminder, reminder from the bank, a reminder from kids, 10% or 20% top-up at the end of the week, 10% or 20% top-up at the start of the week, or a tracking coin? What do you think?
And the results are in!
Any reminder or incentive is better than nothing. This much is obvious. Any condition yielded at least twice the result of just leaving things to take place through motivation.
Post-match top-ups (at the end of the week) are not particularly effective: Unsurprisingly, they proved less effective than the same amount of top-up at the start of the week - loss aversion is a very consistent result across studies. Additionally, a post-match top-up is only about as effective as a reminder.
The difference between a 10% top-up and a 20% top-up, either at the start or end of the week, is not large: In effect, the 20% top-up is literally twice as much as the 10% top-up. But the difference in outcome is minimal. As humans, we often are not really able to grasp percentage differences. What we're concerned about is if there is an increase in the bank account, the exact quantum is not something that we are really able to process specifically.
The reminder from kids worked well: The reminder from their own kids makes the reason for saving even more obvious for the participants. It ties an important emotional attachment to the act of saving. Interestingly, once there is a kid reminder, the top-up, be it at the start or end of the week, doesn't really matter. This gives us a hint that when think about doing something, usually one reason stands out which affects us most.
By far, the coin worked the best: It was more than 50% more effective than any condition or combination of conditions.
Why is this so? Several important reasons. Successfully changing behaviour typically requires the four following factors:
- The outcome is attractive: If the outcome is not attractive to us, then clearly we are not going to pursue it. But how attractive the outcome is also depends how well we can define it. What exactly is it that we want, and why do we want it? The clearer we can define what is it we really want, and why it is important to us, the more committed we will be to attaining the outcome.
However and contrary to popular belief, it is not sufficient to just have an attractive outcome. Knowing our "What" and "Why" is not enough. New Year resolutions are a great example. We really want to fulfil our resolutions, yet we constantly fall short. Climate change is another example. For most of us who accept that man-made climate is serious and will affect our lives in the near future, we often struggle to take climate-friendly action.
This extends to even super-important outcomes like our own health and lives. In other pieces here and here, we covered how people struggle to take medication that will keep them alive, where behavioural scientists had to come up with incentives like a lottery or movies to entice people to save themselves by taking their medicine.
Just an attractive outcome is usually insufficient to change our behaviour. And here is a neat segue into other factors influencing behavioural change.
- The effects are visible and obvious: Imagine if you were trying to get fit in the gym. And after every gym session, you could see significant changes to your body. How exciting and motivating that will be. But more often than not, the changes do not occur so quickly, and we lose our momentum and desire for change.
This is why any intervening condition leads to much better outcomes than the control condition, where participants got no reminders from the bank. Every intervention makes the effort more visible and obvious, reminding the participant that they are doing something, they are making progress.
The coin is especially effective because it makes both process and outcome visible while tying them together. The image on the front and back of the coin provides a visual reminder of the outcome - increased savings. The markings on the coin provide a visible reminder of progress towards this outcome.
- The process is easy: The savings program worked because it was easy to make savings (while very difficult to reverse the process). For the participants, it is merely clicking a button on the phone to transfer funds into their savings account. But to withdraw money, it was deliberately designed where participants had to take a bus to the city centre, fill in a number of forms, and wait for hours before having access to their money.
The easier it is for us to do something, the more likely we will do it.
- The process is satisfying: Unsurprisingly, a text reminder from the kids worked better than a reminder from the bank. It made the participants feel more appreciated, that their kids were aware of the work they were putting in for the family. Naturally, top-ups also work because it is satisfying to see a bigger increase in the bank account.
But why did the coin work so much better than both? It goes back to the markings on the coin - being able to scratch off your own progress is a satisfying feeling. It's like ticking off your to-do list, you are internally congratulating for something you actually did.
While saving money is the example we went through here, the 4 major shapers of behaviour apply to other areas. Behavioural is rarely about knowing what the right thing to do is, but much more about:
Is the outcome attractive, and why is it attractive?
Can the process be made easy and satisfying?
Can progress and outcomes be made visible and obvious?