|
Field
Abstract
|
Before
Equity-Linked Securities (ELS) are high risk structured notes linked to multiple indices, and the recent plunge in Hong Kong H-shares triggered substantial retail losses, underscoring the need for clearer disclosure and bias-mitigating information design. This study examines, in the field, whether such information changes real subscription decisions. The Financial Supervisory Service and one of the largest commercial banks co-design and advise the trial. Implementation takes place at bank branches, and outcome evaluation is conducted at Seoul National University.
Sixty branches are cluster randomized into four arms (A, B, C, D, 15 per arm). Arm A maintains business as usual. Arm B appends an amended profit loss(P&L) chart that front loads loss information and corrects scale distortions. Arm C appends a within ELS comparison that, under identical underlying and fee conditions, displays first barriers alongside the associated coupons, early redemption likelihoods, and loss ranges. Arm D appends a side by side comparison of ELS versus principal protected ELB/DLB. The analysis focuses on KRW-denominated ELS combining KOSPI200, S&P500, and EUROSTOXX50. These are the bank’s highest risk grade products and, under suitability rules, are eligible only for customers with the most aggressive investor profile. Product menus are uniform by date across branches.
The intervention runs for three months. We track behavioral changes in risk-taking, transaction scale, immediate deterrence, post-execution reassessment, and shifts toward safer substitution and diversification. Primary outcomes are the annualized coupon rate (%), the first barrier, and the contract amount (KRW). Pre-specified heterogeneity for seniors (65+, a policy defined vulnerable group) and repeat ELS subscribers identifies for whom the information is most effective. Whereas prior work on structured products has largely relied on lab experiments with synthetic instruments, this regulator bank field RCT observes real products and real purchases, isolating the effect of information framing under a common menu and providing quantitative evidence for reducing high risk choices and supporting standardized disclosure materials.
|
After
Equity-Linked Securities (ELS) are high risk structured notes linked to multiple indices, and the recent plunge in the Hang Seng China Enterprises Index (HSCEI) triggered substantial retail losses, underscoring the need for clearer disclosure and bias mitigating information design. This study examines, in the field, whether such information changes real subscription decisions. The Financial Supervisory Service and one of the largest commercial banks co-design and advise the trial. Implementation takes place at bank branches, and outcome evaluation is conducted at Seoul National University.
Sixty branches are cluster randomized into four arms (A, B, C, D, 15 per arm). Arm A maintains business as usual. Arm B appends an amended profit loss(P&L) chart that front loads loss information and corrects scale distortions. Arm C appends a within ELS comparison that, under identical underlying and fee conditions, displays first barriers alongside the associated coupons, early redemption likelihoods, and loss ranges. Arm D appends a side by side comparison of ELS versus principal protected ELB/DLB. The analysis focuses on KRW-denominated ELS combining KOSPI200, S&P500, and EUROSTOXX50. These are the bank’s highest risk grade products and, under suitability rules, are eligible only for customers with the most aggressive investor profile. Product menus are uniform by date across branches.
The intervention runs for three months. We track behavioral changes in risk-taking, transaction scale, immediate deterrence, post-execution reassessment, and shifts toward safer substitution and diversification. Primary outcomes are the annualized coupon rate (%), the first barrier, and the contract amount (KRW). Pre-specified heterogeneity for seniors (65+, a policy defined vulnerable group) and repeat ELS subscribers identifies for whom the information is most effective. Whereas prior work on structured products has largely relied on lab experiments with synthetic instruments, this regulator bank field RCT observes real products and real purchases, isolating the effect of information framing under a common menu and providing quantitative evidence for reducing high risk choices and supporting standardized disclosure materials.
|