VGSF - WU Vienna - LC

Michaela Pagel, Graduate Business School of Columbia University

online via Microsoft Teams Live Events 17:30 - 18:45

Organizer VGSF

As part of the ser­ies of the "Fin­ance Re­search Sem­inar", VGSF wel­comes Mi­chaela Pa­gel from the Gradu­ate Busi­ness School of Columbia Uni­versity to present her re­search pa­per.
Per­sonal Webpage

Pa­per

Does Sav­ing Cause Bor­row­ing?

We study whether or not nudging in­di­vidu­als to save more has the un­in­ten­ded con­sequence of ad­di­tional bor­row­ing in high-in­terest un­se­cured con­sumer credit. We ana­lyze the ef­fects of a large-s­cale ex­per­i­ment in which 3.1 mil­lion bank cus­tomers were nudged to save more via (bi-)weekly SMS and ATM mes­sages. Us­ing Ma­chine Learn­ing meth­ods for causal in­fer­ence, we build a score to sort in­di­vidu­als ac­cord­ing to their pre­dicted treat­ment ef­fect. We then fo­cus on the in­di­vidu­als in the top quart­ile of the dis­tri­bu­tion of pre­dicted treat­ment ef­fects who have a credit card and were pay­ing in­terest at baseline. Re­l­at­ive to their con­trol, this group in­creased their sav­ings by 5.7% on aver­age or 61.84 USD per month. At the same time, we can rule out in­creases in credit card in­terest lar­ger than 1.25 USD with 95% stat­ist­ical con­fid­ence. We thus es­tim­ate that for every ad­di­tional dol­lar of sav­ings, in­di­vidu­als in­cur less than 2 cents in ad­di­tional bor­row­ing cost. This is a dir­ect test test of the pre­dic­tions of ra­tional co-hold­ing mod­els, and is an im­port­ant res­ult to evalu­ate policy pro­pos­als to in­crease sav­ings via nudges or more force­ful meas­ures.



Back to overview