Investment competence versus confidence: why calibration matters

Imagine two prospective investors who answer the same number of knowledge questions correctly. One feels comfortable asking questions and making a small, considered start. The other assumes that investing is “for people who understand finance properly” and does nothing. A conventional knowledge score treats them as similar. Their journeys are unlikely to be.

This is why investor research benefits from separating competence, what somebody can demonstrate, from confidence, what they believe about their own ability. Neither is a complete measure of readiness. Together, they raise a calibration question; item performance is not a complete measure of competence.

Financial capability is more than knowledge

Financial literacy is sometimes reduced to whether somebody can calculate interest or describe diversification. Objective knowledge matters, but major measurement frameworks are broader. The OECD/INFE survey treats financial literacy as a combination of knowledge, behaviour and attitudes, and also reports financial inclusion, resilience and well-being. Its 2023 international exercise covered 68,826 adults in 39 countries and economies.

The European Commission’s 2023 financial-literacy survey similarly measured both knowledge and behaviours. Across the EU, 18% of respondents received a high overall financial-literacy score, 64% a medium score and 18% a low score. Aggregate categories cannot explain any one person, but they show why “just give people more information” is an incomplete strategy.

Four calibration patterns

Once competence and confidence are separated, four broad patterns become visible:

• Lower competence, lower confidence. The person recognises that they need support. The design challenge is to make learning and a safe first step feel manageable.
• Higher competence, lower confidence. Stronger answers can coexist with unresolved uncertainty about the actual decision. Ask what remains unclear before assuming the person needs encouragement.
• Lower competence, higher confidence. Enthusiasm can run ahead of understanding. Useful support adds a calibration moment without humiliating or obstructing the person.
• Higher competence, higher confidence. Knowledge and self-perception are broadly aligned. The remaining barrier may sit elsewhere: uncertainty, trust, product fit, friction or timing.

These patterns are not diagnoses. They are starting points for asking better questions and designing more relevant experiences.

Why confidence matters for participation

Research has repeatedly linked financial knowledge with participation in financial markets. Van Rooij, Lusardi and Alessie found that financial literacy was associated with stock-market participation in Dutch household data.

More recent work sharpens the picture. Bucher-Koenen, Alessie, Lusardi and van Rooij designed survey modules to distinguish knowledge from confidence. Their model attributed about 30% of the observed gender difference in financial literacy to confidence. This is a study-specific estimate, not the percentage of women or investors who lack confidence. Both measures were associated with participation; this was not a trial of confidence-building support.

This does not mean women are a single behavioural segment or that confidence is the sole explanation for participation gaps. Income, access, experience, product design, social norms, trust and structural conditions all matter. The practical lesson is narrower: a knowledge-only measure can confuse what people know with how certain they feel about answering.

How to measure without pretending certainty

A responsible assessment should avoid asking one question to do several jobs. At minimum:

1. Measure self-perceived knowledge or confidence directly.
2. Use separate, objective knowledge questions appropriate to the population.
3. Compare the signals using a documented rule.
4. Describe the result as a possible calibration gap, not a fixed personal trait.
5. Review results across projects for unexpected demographic or contextual skews.

At U Impact, fixed rules make a comparison reproducible for the same answers, items and scoring version. That does not establish its validity. Confidence should refer to the concept being tested, and scales need a documented comparison rule. Review item interpretation and group differences before using a result to assign support.

From signal to better design

The value of identifying a gap lies in what a team can test next.

When confidence trails competence

Ask about the decision before offering reassurance. A correct answer about diversification does not resolve uncertainty about fees, financial capacity or a sustainability claim. Explain the concept the person demonstrated, identify what remains open, then test whether the proposed support helps them understand their options.

When confidence runs ahead of competence

Add a respectful knowledge check, make downside scenarios concrete and create a pause before commitment. The goal is informed agency, not friction for its own sake.

When both appear aligned

Look beyond education. The limiting factor may be ambiguity, trust, choice overload or a poor hand-off between interest and action.