Does Matching Contribution incentivize informal workers to participate in retirement saving plans? A Randomized Evaluation interacted with a Natural Experiment

Abstract

We conducted a large field experiment in Peru on informal workers and studied whether offering them a matching contribution raise participation and con- tributions in their Individual Retirement Accounts. We had three groups: a control group receiving no match, and two treatments groups receiving 50 and 100 percent match, respectively. Additionally, due to the time span, we can also analyze the difference responses between pre and during Covid-19. The results were as follows. First, the match incentive increases participation. Workers in the 50 and 100 per- cent match groups show participation rates of 5.2 and 6.5 p.p. higher than workers in the control group, respectively. The participation effect is also present pre Covid- 19 and disappears during it. Second, the 100 percent match incentive was the only effective in increasing savings among all individuals (1.4 p.p.), pre (2.3 p.p.) and during Covid-19 (0.97 p.p.). This effect still presents in LATE specification with higher p.p. Third, 100% match was again the only effective to make contribute more than once, in the full sample (1.2 p.p.), and pre Covid-19 (2.7 p.p.), including LATE specification (full sample 5.6 p.p.; pre Covid-19 8.2 p.p.). Fourth the 50 percent match is not effective in raising contribution in any specification.