Terug

A randomized controlled trial on the effectiveness of mandatory investment information

When securities are offered to the public or admitted to trading on a regulated market, a prospectus often needs to be published. For retail investors, this disclosure also has to be summarized in a summary prospectus and in a Key Information Document (KID).

Because the AFM aspires to evidence-based policy for fair and transparent financial markets, we assessed the effectiveness of the summary prospectus, the KID, and proposed combination of these two documents. The AFM has a dedicated, online Consumer Panel that allows for a randomized controlled trial. Are retail investors able to identify and choose the best investment, using different types of mandatory disclosure?

Purple dominates

We constructed three different investment products: “Eurozone Bond Series Purple”, “Eurozone Bond Series Yellow” and “Eurozone Bond Series Blue”, where the Purple-bond was always better (or at least equal) to the other two bonds; higher yield, lower risk and lower costs. The Blue-bond was also dominated by the Yellow-bond. The task at hand for the respondents was to distribute €10,000 over these three investments. The correct answer would be to allocate the full €10,000 to the Purple bond.

The screenshot below shows the webpage where respondents could access the documents and allocate their €10,000. We include this to give a feel for the task that was asked of the Dutch respondents. Superimposed in a green box are the differing characteristics of the three bond offerings. All other texts in the information documents was equal between the different Eurozone bonds.

consumentengedrag-art-prospectus-kid-figuur

Experimental set-up & response

Respondents were randomly assigned to one of three disclosure groups. In one experimental group, they received the information as summary prospectuses (four pages), one each for the purple, yellow, and blue bond. A second group could assess the different investments using three Key Information Documents (each KID came to three pages), and the third experimental group was provided with a combination of a summary prospectus and a KID, amounting to documents of six pages. The three bonds were shown in random order for each respondent.

Out of the N=1597 people that were invited, 729 (46%) started the questionnaire, N=157 consumers (22%) were excluded because they were not retail investors or had never invested on the stock market. Out of the remaining 572 eligible participants, N=384 completed the full survey. The attrition-rate of 33% was much higher than in other surveys in this panel (usually non-completion is <5%), testament to the high demand this task placed on respondents. Participants had to read three documents that each consisted of between three and six pages. There were no significant differences in abandonment between experimental groups.

Subjective ratings

The summary prospectus was perceived as significantly less useful and less clear than the KID or the combination document. Perceived length was similar for the three types of documents. The summary prospectus also scored significantly lower than the other two on the statement: “These documents enabled me to make a good choice for distributing €10,000 over the three investment products”.

Best decisions with KID

The amount invested in the purple bond gives a measure of objectively correct a retail investor allocates his €10,000. Over one third (34%) of the respondents that read the information as Key Information Documents, correctly invested everything in the dominating purple bond, for the combined disclosure this was 31% and for the summary prospectus 24%. On average, the KID-group invested €5,992 in the best bond, significantly more than the €4,872 for the summary prospectus group. The combination group on average invested €5,701 in the purple bond.

The graph below shows the cumulative distribution of investments for the three types of disclosure. In all groups, a little more than one in ten did not allocate anything to the best bond. In general, the more to the right the curve is for a specific type of disclosure, the objectively better the decision by the respondents.

consumentengedrag-art-prospectus-kid-grafiek

For an ordinal regression, we created five bins, shown in the graph above. Running a full model, we see a significant better asset allocation for the KID and the combination, compared to the summary prospectus. Also, the more time taken for the task, the better the allocation. People spending more time reading the documents, allocated more money to the dominating investment. Respondents who are execution only investors (compared to investing with advice or management) made objectively better investment decisions in this task. The order in which the three different bonds were presented, had no significant effect on the outcome. Age, gender, education, number of years investment experience, or amount of assets invested in real life all had not significant explanatory power in determining the investment decision in this research.

Conclusion

Using realistic types of disclosure in an online experiment, we show that Dutch retail investors objectively allocate assets better using Key Information Documents (KID) than with the summary prospectus.

We could assess the objectively best choice, because in our controlled setting, there was one offering (purple) that dominated the other two bonds; costs and risks were lower or equal, and yields were equal or higher. Obviously, the real world is much more complex and less unequivocal. However, randomized controlled trials such as this research allow for evidence based policy and mandating disclosure that allows retail investors to make better decisions.

In general, we could conclude that less is more. This is an important lesson for (European) policy makers designing new or improved mandated disclosure. A larger stack of documents does not necessarily mean that retail investors will be able to make better investment decisions, or more broadly: more information might even lead to less optimal financial decisions by consumers.

Informatie delen

Delen via: deel
Alle onderwerpen