Superfreakonomics

Superfreakonomics reveals how you can find non-obvious solutions to tricky problems by focusing on raw, hard data and thinking like an economist, which will get you closer to the truth than everyone else.

Even though we can’t see into peoples’ heads and just look at what exactly makes them tick, we’re still driven by the power of incentives all the time. Governments, companies, schools, even just other people constantly try to get us to do things by dangling certain rewards in front of us.

The idea is simple enough: You promise someone reward B for performing action A and hope that everyone in your target group shows the desired behavior.

However, there’s a hidden force at play here: the law of unintended consequences. Dubner and Levitt have coined this phrase to describe the behaviors that occur after giving an incentives that weren’t planned.

For example, in Germany, the government keeps trying to get people to produce less waste with fun ideas like picking up trash only once every three months, downsizing trash cans or introducing volume-based fees. So far, all of these ideas have backfired horribly, leading only to creative ideas on how to avoid the new systems, for example by dumping trash in the woods or flushing food down the toilet.

But even the incentives that do work won’t do so for everyone, and all of them will have some side effects. You might get your kids to do their dishes if you promise them $1 for every cleaned plate, but that might also lead them to clean them even when they aren’t dirty or expect money for other household chores.

Don’t forget: all incentives have intended and unintended consequences!

What do scientists do if they can’t find the solution to a problem? They collect data! Data always helps you find a solution, but it might not do so in the way that you think. For especially difficult problems, the solution often lies one level above the realm of the issue itself. A single data point, an extreme value or outlier, or the data points you thought you’d collect, but didn’t, often tell a much more revealing story than the “normal” data. This is related to a phenomenon called omitted variable bias, which means you’ve forgotten to even include one of the most important factors in your analysis. For example, there’s a correlation between a country’s chocolate consumption and its number of Nobel prize laureates, but that doesn’t mean eating chocolate makes you more likely to win a Nobel prize. It just so happens that wealthier countries with a higher level of education also spend more money on luxury foods. Similarly, it’s sometimes easier to find solutions that prevent…