BEHAVIORAL FINANCIAL PRODUCT DEVELOPMENT: Primer, Progress, Frontiers Jonathan Zinman Dartmouth College and IPA’s U.S. Household Finance Initiative (also J-PAL, NBER, etc.) A. Primer Symptoms and Causes: Behavioral (Household Finance) Economics 101 Applications: Markets Your employees Scope today More about product development Much less about marketing, persuasion Tmrw at Behavioral Finance Forum Financial Illness: Symptoms Widespread low financial resiliency Little savings for many households High debt reliance: expensive High “money on the table” Poor shopping, mediocre management Low financial sophistication Problems => Opportunities A. Primer Financial Illness: Causes (Behavioral Economics 101a) #1 Cognitive biases that stack deck toward spending/borrowing, away from saving/accumulating In preferences: costly self-control, loss-aversion In expectations: “things will get better” (or at least not worse) In price perceptions Underestimation of compound interest Underestimation of borrowing costs Limited attention A. Primer Financial Illness: Causes (Behavioral Economics 101b) #2 Mistakes borne of misguided heuristics, other cognitive limitations Information/choice overload Anchoring Low (financial) literacy, numeracy A. Primer Financial Illness: Causes (Behavioral Economics 101c) #3 Limited opportunities for learning … on high-stakes decisions Mortgage/house Job Marriage Car (and financing it) Even high-frequency decisions can have uncertain long-run implications Credit card use (what’s right debt load for me/my family)? Changing life circumstances creates moving targets A. Primer Financial Illness: Causes (Behavioral Economics 101d) #4 Markets sometimes exacerbate consumers’ cognitive “bugs” Advice markets are a mess and limited in scope Who covers the household balance sheet? For the mass market? Price competition in product markets helps, but only partly A. Primer Taxonomy of Behavioral Factors (see DellaVigna JEL) Biases in expectations • Overconfidence • Over-optimism “Cross-cutting” biases/heuristics/ limitations • Anchoring • Limited attention • Innumeracy Biases in price perceptions/valuation • Exponential growth bias • Anchoring Biases in preferences • Time-inconsistency • Loss aversion A. Primer DECISION Alternate Taxonomy Preferences Biases/limitations in cognition that affect perceptions about how to maximize utility subject to constraints (And hence affect other key parameters we might model: expectations, prices, transaction costs…) DECISION A. Primer B. Progress By way of examples: 10 pilots (alpha-/beta-tests) All with “retail” financial service firms (D2C) “Wholesale”– particularly through employers– is another promising channel (E2C). A la HelloWallet Completed Pilot 1: Performance Bond for Smoking Cessation “Put Your Money Where Your Butt Is!” Bank offered in Philippines Bank account offered to smokers “Deposit money you were spending on cigarettes” Agree to a urine test in 6 months If nicotine free: get your money back (no interest) If not nicotine free: your money goes off to charity Result: 11% opened an account 30 percentage point increase in smoking cessation Example of a “commitment contract” B. Progress Commitment Options Commitment contract = voluntary restriction, or self-provided added incentive, in service of goal attainment Performance bonds Liquidity restrictions (e.g., spending limits) “Cut me offs” Peer support/social reputation (stickk.com, other models) (Also automation; e.g., auto-deductions) B. Progress Completed Pilot 2: Messaging as a Product Feature SMS Reminders for Goal Attainment Pilot With savings account holders at 3 different banks in 3 different countries Reminders raised balances by 6% Now extending to debt reduction, budgeting, and planning goals B. Progress Completed Pilot 3: Borrow Less Tomorrow (BoLT) Behavioral Kitchen Sink for Debt Reduction Decision Aid Escalating Repayments Peer Support Reminders/Feedback (Monitor progress using credit reports) Save More Tomorrow™ as a guide High-touch pilot test at free tax prep site in Tulsa 41% take-up rate 37% plan escalating repayments; 26% enlist peer supporters 51% on-schedule after 12 months (is this high or low??) B. Progress Pilots 4-10: Financial Products Innovation Fund “The Financial Products Innovation Fund was created as a joint effort between IPA’s US Household Finance Initiative and the Ford Foundation to support the development of scalable, market-tested products that help households make better financial decisions, escape cycles of debt, build assets and achieve financial resiliency.” • Emphasis on product development informed by behavioral research • Theory, evidence, principles • Competitive call for proposals launched in August 2011 • Seven projects awarded funding in January 2012 • Product development and “alpha-testing” for feasibility and level of demand • Will also be doing some analysis of demand determinants • Pilot tests currently underway (Apr – Dec 2012); results by May 2013 B. Progress “Pay Yourself Back” – Two Pilots • Problem: Hard to get started saving • Solution: Seamless conversion of loan/DMP payments to savings once loan/DMP is paid off • Behavioral approaches: Harness habit formation, redirect mental accounting, easy on-ramp (use existing accounts) • Key Features: Upfront commitment, back-end automation • Take-up rate of about 10% B. Progress Pay Back Yourself– “Get Saving!” B. Progress “Frictionless Savings” – Two Pilots • Problem: Easy to spend on impulse, but not to save (on impulse) • Solution: Clear path to saving at times when you are most liquid • For example: at the check cashing window • Approaches: Redirect impulsivity, meet people where they prefer to conduct business, streamline bank account sign-up • Key Features: MicroBranch: Upfront commitment, back-end automation; RiteCheck: “impulse saving” • Target Market: Low-income check cashing customers in San Jose & New York City • Take-up rate: about 25%. B. Progress “Frictionless Savings” – Two Pilots B. Progress “TGIF: The Goal is Freedom” Loan • Problem: Small dollar loans are expensive, and have high default rates • Solution: Borrowing accepting 5-day delay in disbursement gets 28% - 69% discount • (Sister product: emergency line of credit with “frictionful” drawdowns) • Approaches: Incentivize planning, risk screening-by-sorting • Key Features: Delayed disbursement, frame interest as savings to incent improved financial planning; (voluntary liquidity restriction) • Target Market: Roanoke credit union members, 61% designated low-income • Take-up rate: about 40% B. Progress “The Trust Card” Credit Card • Problem: Yield-maximizing strategy for many households is to pay down highinterest debt, but many only make minimum payments each month • Solution: Behavioral Credit Card for debt consolidation • Approaches: Decision aid; default option; commitment options • Key Features: default to accelerated payment plan; built-in planning tool; restricted access to liquidity going forward; option of further voluntary restriction • Target Market: Low-income credit union members in Washington Heights, NYC • Takeup rate: About 17% B. Progress “The Trust Card” Credit Card B. Progress “Now and Later Account” Restricted Withdrawal Account • Problem: The temptation of lump sums (some jobs, many tax refunds) • Solution: Voluntarily restrict own access to lump sums • Approach: Decision aid, commitment, and automation bundled with savings account • Features: Budget commitment to withdrawal schedule • Target Market: Students via Single Stop USA’s Community College Initiative (smooth disbursement of student loan payments) • Expected Launch: August 2013 B. Progress “The Now & Later Account” (See also “Aid Like a Paycheck”) The Now&Later Account B. Progress Next Steps • Pilot products: Scale, Refine, Evaluate •With randomized-control testing • New ideas: Develop and launch several new alpha-tests B. Progress Progress: How We Make it A Virtuous Cycle: R Test D B. Progress C. Some Frontiers Innovations in small-dollar space Documentation Peer referrals Commitment options Loan shopping Cracking willingess-to-pay key Transparent pricing <-> Scalability Personal financial management Content, take-up, engagement Summing Up Behavioral insights can help guide how you help people (customers, employees) deal with money Product Design Marketing Pricing Process (e.g., “on-ramps”) Communication/Engagement Etc.