Framing bias

Offer people the same choice twice, worded differently, and many will choose differently. That is framing bias.

The workbook defines it as an information-processing bias in which a person answers a question differently based on the way it is asked (framed). An investor's choices are influenced by how information or facts are presented, not only by the facts themselves.

It is one of the cognitive errors — a result of faulty reasoning rather than feelings — alongside mental accounting and anchoring.

Two types the workbook names (Chapter 14, section 14.3.2):

In both pairs the outcomes are identical, yet people mostly opted for the second option.

Link to prospect theory. Prospect theory considers how prospects are perceived based on their framing; its first phase is the one in which prospects are framed.

Where it bites in portfolio management:

1. Risk tolerance questionnaires. Investors have to recognise this bias when filling them, because the answers greatly influence asset allocation.
2. Performance evaluation. Decision frames may lead to sub-optimal judgements when a portfolio's performance is reviewed.

Illustrative figures. A PMS client, Mr. Bose, has ₹1 crore to allocate. His risk questionnaire asks one question in two ways.

Version A: "This portfolio has a 90% chance of not losing money in any year. Comfortable?" — He ticks Yes.

Version B: "This portfolio has a 10% chance of losing money in any year. Comfortable?" — He ticks No.

The two statements describe the same portfolio (90% + 10% = 100%). Yet Version A would place him in a higher-risk profile — say ₹70 lakh in equity — and Version B in a lower one — say ₹40 lakh in equity. A ₹30 lakh difference in equity exposure came from wording alone.

A careful manager asks both versions, notices the inconsistency, and discusses the actual numbers with Mr. Bose. The same applies at review time: "we beat the index by 2%" and "we lost 8% while the index lost 10%" can describe the same year.

Chapter 14 (Behavioural Finance), section 14.3.2, covers framing among the cognitive errors, with the risky-choice and attribute-framing examples and the warnings about risk tolerance questionnaires and performance evaluation. Chapter 9's risk profiling (section 9.3.3) is where that warning applies in practice.

• Framing is a cognitive error (information-processing bias), not an emotional bias.
• The outcomes in a framing pair are identical. If the underlying numbers differ, it is not a framing example.
• Two named types: risky choice framing (lives lost vs saved) and attribute framing (0% fat vs 100% fat free). Do not swap the examples.
• Framing is the first phase of prospect theory; evaluation is the second.
• Do not confuse with anchoring. Framing is about how a question is presented; anchoring is about over-relying on an initial piece of information.