Child and Youth Financial Inclusion. Dean Karlan Yale University Innovations for Poverty Action M.I.T. Jameel Poverty Action Lab

Transcription

1 Child and Youth Financial Inclusion Dean Karlan Yale University Innovations for Poverty Action M.I.T. Jameel Poverty Action Lab

2 Takeaways Not going to hear do this and all will be solved If problems were that easy, we wouldn t have such gatherings Two main hoped-for takeaways: Get ideas on best ideas, inspired by theory, tested in practice Answer is often in the process The how is as important as the what Lesson for the high-level: Setup the right process for a learning program

3 When to translate research to action? How do you translate lessons from place A to B? Good theory Good evidence

4 Session Overview 1. Policy Challenges Youth unemployment School financing for households 2. Case studies 3. Conclusions & Way Forward

5 Big Picture: Two Policy Challenges Youth unemployment Regulatory Labor market frictions Human capital: vocational, basic skills and character Access to finance (for both education & entrepreneurship) School finance for households Take Egypt as an example Financing for tutoring, school supplies, fees, exams. Low access to savings Similar issues around the world.

6 Translating research to action Step One: What is the theory of change? Economic failures Perfect information?? High search costs or information acquisition costs Zero transaction costs?? High transaction costs for bank accounts Both? Behavioral failures Disciplines, time inconsistency, attention Youth: Perhaps behavioral failures can be addressed, but before habits and behavioral patterns are formed? Commitment: how to structure?

7 Translating research to action: Step two: Tinkering One key lesson: Subtle variation: savings account withdrawal in cash or voucher? Tinkering matters! Sometimes the idea isn t what matters so much. But the how is what matters very much.

8 New Evidence in Child and Youth Finance THREE EXAMPLES

9 SaLSa

10 The Policy Problem Youth (16-28) Forming long term habits Entrepreneurship/labor: need money to make money, facilitate savings The specific challenge: Education alone sufficient? Account access alone sufficient? Or need both?

11 Access to Accounts Theory of Change Access to Savings Accounts Exposure to services Awareness of what s available Trust services Improved savings behavior Variations

12 Financial Education Theory of Change Financial Education Precise financial knowledge Understand financial self Trust financial services Informed financial decisions Variations

13 Three Evaluations 1. SaLSa Uganda Relative impacts of 2 programs: Financial education and Group savings account Church of Uganda youth groups (16 28 years) May 2010 August 2011 Co-investigators: Julian Jamison (Federal Reserve) and Jonathan Zinman (Dartmouth)

14 Research Questions Intersection of Programs Financial Education BOTH Savings Accounts

15 SaLSA: Financial Knowledge Index Std Dev Change 0.15 *** *** FE + Account Account Only FE Only Control Treatment Groups

16 SaLSA: Total Self Reported Savings Percent Change ** *** FE + Account Account Only FE Only Control Treatment Groups '000s UGX (winsor 5%)

17 SaLSa: Final Balance of Deposits, Admin Percent Change ** FE + Account Account Only Account Treatment Groups '000s UGX (winsor 5%)

18 SaLSa: Total Earnings Percent Change ** * * FE + Account Account Only FE Only Control Treatment Groups '000s UGX (winsor 5%)

19 Super Savers

20 Policy Challenge Households report lacking funds for education expenses Many choices: Subsidize school expenses more Budget What to subsidize? Improve school quality, training, etc. Tinkering matters!

21 Basic theory Basic problem: no cash for school expenses Yet stated desire to have savings We all make plans And fail And regret And do it again

22 Three Evaluations Super Savers Uganda Variations on savings accounts: Cash payout vs Voucher payout Variation on parent participation Primary schools (10 15 years) September 2009 to September 2011 Co-investigator: Leigh Linden (University of Texas Austin)

23 Results Outcomes of Interest Savings behavior Self-reported behavior Administrative data (Super Savers) Also interested in process outcomes, to understand Intermediate outcomes/process outcomes Results on savings

24 Super Savers: Self Reported Savings Percent Change ** Cash w/ Parent Outreach Voucher w/ Parent Outreach Cash w/o Parent Outreach Voucher w/o Parent Outreach Control Treatment Groups '000s UGX (winsor 1%)

25 Super Savers: 2011 School Supplies Index Percent Change 300 ** Cash w/ Parent Outreach Voucher w/ Parent Outreach Cash w/o Parent Outreach Voucher w/o Parent Outreach Control Treatment Groups From classroom visits

26 Super Savers: Normalized Test Scores Std Dev Change ** Cash w/ Parent Outreach Voucher w/ Parent Outreach Cash w/o Parent Outreach Voucher w/o Parent Outreach Control Treatment Groups

27 Aflatoun

28 Three Evaluations 3. Aflatoun Ghana Variations on financial education: Social & financial education vs Financial education only With school savings club Primary and junior secondary schools (6 14 years) August 2010 to July 2011 Co-investigators: Jim Berry (Cornell) and Menno Pradhan (University of Amsterdam)

29 Aflatoun: Savings Attitude Index Std Dev Change ** Aflatoun HMB Control Treatments

30 Aflatoun: Financial Literacy Index Std Dev Change Aflatoun HMB Control Treatments

31 Aflatoun: Savings Behavior Index Std Dev Change *** 0.2 ** Aflatoun HMB Control Treatments

32 Aflatoun: Work Index Std Dev Change * Aflatoun HMB Control Treaments

33 New Evidence in Child and Youth Finance CLOSING

34 Closing Takeaways Youth: Access to savings and financial education positive change Complements? If not, tradeoffs! Which is cheaper? Inconclusive results on financial knowledge Impacts non-financial outcomes too Education spending Participation in workforce Tinkering matters.

35 An Opportunity: Role of NGOs in Financial Access Innovation Unreached Too young Too poor Too rural Trust

36 Closing Way Forward Guiding principles Access matters Make it easy Evidence from elsewhere Context appropriate? Politically viable? Evidence from within Does it work? How to make it work best?

37 Thank you!