For many people, saving money isn’t just difficult; it’s a foreign concept. A recent study found that 58% of Americans do not have a formal retirement plan in place.¹ Why is even thinking about saving money so daunting to so many of us?
We spoke with Dan Ariely, professor of psychology and behavioral economics at Duke University in Durham, North Carolina. He says that many people have difficulty saving their income because our minds and our environments are not naturally suited to thinking about money in the long term. In fact, our minds are not very good at thinking about the concept of money at all. To make it easier, he says, we need to change our environments in such a way that saving money happens automatically and we never even have to think about it.
More insight from Dan Ariely:
- Dan Ariely.com
- TED Talks page
- Dan Ariely’s Predictably Irrational page at MIT
- Dan Ariely’s previous commentaries on Need to Know
- Twitter profile: @danariely
¹“According to Deloitte’s Retirement Survey, a majority of Americans — 58 percent — do not have a formal retirement savings and income plan in place.”
One of the biggest lessons in social science over the last 40 years is that we are creatures of our environment. So the decisions we make are actually an outcome of the environment we’re being placed in. And if the environment tempts us, for example, think about every morning you come to your office, and your desk is covered with doughnuts. That’s not going to have a good– effect on you.
Saving money is probably one of the toughest things that people can– can do. And– and the re– there are two reasons for that. Is– the first one is between now and later. And here’s a general way to think about this trade-off. Imagine I ask you, “What would you rather have– a half a box of chocolate right now, or a full box of chocolate in a week?” Most people said half a box of chocolate right now, full in a week– give me the half now. But imagine we push the choice to the future. And we said, “What would you rather have: half of box of chocolate in a year, or a full box of chocolate in a year and a week?” It’s the same trade-off. It’s asking you whether you’re willing to wait another week for another half a box of chocolate. But now, most people say, “I’ll wait another week for another half a box of chocolate.”
The second problem with money is that money is abstract. In fact, money is incredibly difficult to think about. Every time you think about a cup of coffee, you should be thinking about, “What am I giving up for this cup of coffee?” But that’s a very hard computation, right? What– what exactly are you giving up?
So it’s now versus later, concrete versus abstract. And because of that, it’s incredibly hard for us to save for long term. It’s incredibly hard for us to think about the value of money at retirement. And it’s difficult for everybody. So what do we do when something is so difficult for us to do naturally? One approach you say educate people. Give people all the information, teach them about how to compute– compound interest and the value of money, and so on. Turns out there’s no evidence that this is working. Instead what we need to do is we need to re-engineer people’s environments so it’s easy to save and you don’t have to think about it.
I think that the best example of a public policy that has been informed by social science is the default of opting out from retirement savings. So it used to be that you came to a company and you started working and they said, “Would you like to save for retirement in a 401(k) plan?” And you said yes or no and how much. Now they said, “You’re in the program. Do you want to get out?” And that has been a very successful plan because of what we call opt in and opt out that going into a program we need effort, decisions, and so on. So we’re less likely to do it. But if we’re in the program already by default, coming out requires effort and attention and so on. So we’re less likely to do it and we’re likely to keep with the plan.
So what we need to do is we need to think carefully about the environment and we need to design an environment that is compatible with the way we process information. We really need to think about what we’re capable and not capable. And we need to design the environment that is tailored to our capabilities and not over-assuming that we can compute something that we really can’t.