I posted recently about work incentives and the difficult relationship between effort and money. A lot of traditional economic models assume that increasing monetary rewards increases the effort that workers choose to put in, and therefore performance. But sometimes, paying people to do something makes the task less enjoyable.
So, give some kiddies from Coshocton, Ohio some money to do well at school and what happens? They do better in Maths tests, but not in reading, social science or science. And these were big incentives for a small child: up to $100. More than I used to have in my piggy bank when I was at primary school.
Why do the incentives work for Maths, but not other subjects? I think it’s tempting to read too much into this result. For some tasks, there is a closer link between effort and performance. Take, for example, stuffing leaflets in envelopes versus writing a book. Spending more time stuffing envelopes equals more stuffed envelopes. But spending more time trying to write a book doesn’t always equal a longer or higher quality book: sometimes the ideas just don’t flow. Something like this probably applies for learning as well: in some cases, you can step up the effort and performance follows. Other times, it just doesn’t. But it’s tricky to tell exactly what’s driving the results of the Coshocton kids’ study without knowing how the subjects are taught and how the tests are structured.
Another intriguing aspect of this study is how the kids are paid: in “Coshocton Children’s Bucks” – gift certificates redeemable in local shops. The reason for not using cash was to avoid the potential for unscrupulous parents to appropriate the proceeds for their own selfish ends. I wasn’t sure whether to be amused or appalled at this lack of faith in humanity.
The link to the paper is here: http://papers.nber.org/papers/w16333