Testing the Effectiveness of Consumer Financial Disclosure: Experimental Evidence From - PowerPoint Presentation

1. Testing the Effectiveness of Consumer Financial Disclosure: Experimental Evidence From Savings AccountsPaul AdamsStefan HuntRedis ZaliauskasFinancial Conduct AuthorityChristopher PalmerUC BerkeleyCFPB Research ConferenceDecember 2016The views expressed herein do not necessarily represent the official position of the FCA.1

2. Motivation behind financial disclosureIn theory, so long as consumers and firms are completely informed, the market “gets it right”(in the sense that there’s no way to improve someone’s position without making someone else worse off)But being “completely informed” is a myth:Goods and services are complex and costly to evaluate.Popular solution: policy often mandates disclosing information, firms often want liability indemnification.Surgeon general warnings, warranties, terms and conditions, truth-in-lending law disclosures, …Unintended consequence: an abundance of fine print2

3. Research QuestionsHow much does the design/presentation of disclosure matter?Can optimized disclosure improve consumer outcomes?Or are there natural limits to the ability of consumer financial disclosure to protect consumers?3

4. Why savings accounts are the ideal setting for testing these questionsProblem of inferring effectiveness of disclosure: need to take a stand on what the “right” decision is.“The Joint Hypothesis Problem” in Asset PricingOne dimensional differentiation: interest rateNot entirely true: branch network, app quality, bank reputation, synergies across accounts: will address these“Front-book/back-book” pricing models4

5. Disclosure’s uphill battle:consumer choice is stickyIf consumers have ample options and information, we’re generally not worried about firms delivering substandard goods and services because people will simply switch providers to one that meets their standards.But consumer inertia can get in the way“Switching costs” are any time or stress or hassle that consumers must incur to reoptimize their decisions.Long literature in economics, marketing, psychologyThree flavors of switching costs:Uninformed – lacking information about pros of switchingInertia – have information but reluctant to switch because the process is too much of a pain given the size of the gains.Inattention – not paying attention to issue one way or the other5

6. UK Savings Market Background62 Same products sold for different prices1 Large and important market3 Simple environment93%>$1 trillion80% not switched in last three years60%30%0%3.0%1.5%1.0%% of balancesInterest rate<2 years 2-5 years >5 yearsΔ% clear benefitscontrol forbrandpreferencelow switching cost

7. 5 randomized controlled trials (RCTs)Reverse pageFront pageSMS reminderEmail and SMS reminderReturn switching formSwitching boxReminders>134,000 customers in totalMedian gain £32/year, average gain £127/year.7SampleTreatment arms13,000463,00044,000130,000424,0002+ survey+ survey

8. Example Disclosure Design: Reverse Page8

9. Front-page letter call to action9

10. Front-page switching box: Best rates10

11. Return switching formControl: Treatment:11

12. Return switching form12

13. Email and SMS remindersReference to best comparable account with the provider13

14. RCT DesignFive different providers each having different customer mix.Each randomized customers into treatment and control groups and delivered corresponding disclosures.Follow-up survey: Followed up a limited set of customers with survey.Strength: can compare results across providers to establish external validityWeakness: hard to compare treatment effects across providers given different customer mix14

15. DataRich administrative data on consumersAverage age: 54Has checking account with same provider: 55%Average account balance: £17,147 (median ~£5k) Account age: 8.8 yearsShare using online banking: 47%Significant variation across providers9-90% use online bankingAverage account 1-16 years oldAverage consumer age 42-646-97% with checking account at same provider£1k-£12k median balances15

16. Outcomes: measuring effectivenessProduct differentiation besides interest rates makes classification of “wrong” decisions problematicMotivates focus on “internal” switching: can take a strong standHard to rationalize preference to stay with dominated easy-access savings product at the same bank (apart from switching costs, information, inattention, and inertia)16

17. Results: design matters but effects modest 17Rate cutRate cutEnd of bonusNo changeNo change

18. Email and SMS reminders impacted internal switching only18

19. Return switching form inducedonly internal switching19

20. Who responds best to disclosure?Perhaps the gains just aren’t enough to care about => can restrict to large balancesPerhaps I like my bank: have my checking account there, trust the brand, find the ATMs convenient. => can restrict to internal switching only.Other controls: age, gender, balance, number of products with same provider, account age, etc.20

21. Lessons for Disclosure DesignBurying the disclosure on last page: little effectGraphical depiction of disclosure: little effectProcess improvements: facilitating internal switching had strongest effectEffects strongest when tied to a nearby salient event: impending/recent rate decrease21

22. Survey resultsIn the survey we measured:recall (interest rate, interest rate change, and communications received)intermediate actions of customers (shopping around, and the time it took)subjective evaluations (satisfaction with the individual decision taken)Consumers generally overestimate interest ratesMany can’t recall getting or noticing disclosureMost expected switching process to be more onerous than it turned out to be22

23. ConcernsIs there some optimal form of disclosure that does overcome switching costs/inattention?Inattention probably rational given the importance of average consumer disclosure. How to avoid “alarm fatigue” in consumer protection?Suspiciously off-equilibrium to have firm do this?Firms’ equilibrium response to disclosureObfuscationChanging pricing models23

24. ConclusionWe tested targeted information disclosure and switching process simplification for a large sample of customers.Design matters, but even best designs have modest effectsEven for people that have the option to easily switch to a better account with the same provider.Even for people with large balances and a lot to gain from switching.Little evidence regulators could mandate some magic optimal design that facilitates attentiveness and action.Faced with mandated disclosure, infinitely many ways for firms to blunt its effects…Calling into question the policy reliance on disclosure.Regulators should consider using nudges and harder interventions in addition to or sometimes instead of disclosure.24