How peer pressure can help you save

What do you think would make you more likely to save—being paid a higher interest rate or the social pressure of making a public commitment to save more and then having your peers monitor your progress?  A new paper from Felipe Kast and Dina Pomeranz (hat tip to the CMF folks for blogging it first) finds that while both strategies help to increase savings, the peer commitment mechanism comes out on top.

Kast and Pomeranz set up a randomized control trial using groups of Chilean microcredit clients—2,682 individuals in total. Each group was randomly assigned to be offered one of three types of savings accounts provided by the Banco de Chile. The first was a basic account with a 0.3% real interest rate. The second was the same account but with a 5% real interest rate. The third and final condition was the same basic account with a peer commitment feature, whereby group members had the option to publicly announce their savings goal at the first meeting, and then in subsequent meetings to discuss who complied with their goals and who did not.

I should note that all three types of accounts were attractive to begin with compared to other options in the market: they had no maintenance fees, no minimum balance and came with an offer to go as a group to the bank to open the account. But the study results show a dramatic difference in the effect of the three different types of accounts.

Individuals assigned to the peer-based commitment device made about three times more deposits and had an average balance in their accounts that was 65% higher than the balance of those with the regular savings account without the peer mechanism.  Kast and Pomeranz did the math and estimated that the effect of the peer mechanism corresponds to an interest rate increase of approximately 6.5 percentage points. This is striking in and of itself, but it was another finding that jumped out at me. The researchers found that the peer effects were stronger for individuals who had more confidence in their ability of following through with their plans, i.e., those who felt more certain that they could achieve their stated savings goals. The reasoning here is that people who think they have fewer self-control problems than others feel most encouraged by observing their peers’ behaviors in any situation.  For example, when they see someone succeed in reaching a goal, they think they can do it just as well. On the flip side, when they see someone fail, they’re not as discouraged because they think they have better self-control skills to begin with.

These results bring a lot to bear on the design of new savings products that have the goal of increasing savings. However, it also seems to me that this kind of nonlinear result for the peer commitment mechanism could have important implications for certain policy initiatives that put people into groups for training or other programs.