Behavioural biases and retail investor responses to perceived currency volatility in the Indian mid-cap equity market

Abstract Retail participation in India’s mid-cap equity segment has grown quickly, and mid-cap valuations are widely regarded as more sensitive to macroeconomic shocks — including exchange-rate movements — than large-cap valuations. This study examines how three behavioural biases (herding, overconfidence, and the availability heuristic) relate to mid-cap investment decisions among Indian retail investors under conditions of perceived rupee volatility, and tests whether financial literacy and social-media reliance are associated with these biases. Cross-sectional survey data were collected from active retail investors ( N = 519 valid responses after listwise deletion) and analysed using confirmatory factor analysis and a structural path model, with 1,000-sample non-parametric bootstrapping used to obtain confidence intervals for all indirect (mediation) effects. Social-media reliance was positively associated with all three biases (herding, overconfidence, and the availability heuristic), while financial literacy was negatively associated with herding and the availability heuristic but was not significantly associated with overconfidence — a pattern consistent with, though not proof of, a “competence effect” in which self-assessed knowledge does not by itself reduce overconfidence. Herding, overconfidence, and the availability heuristic were each positively associated with investment-decision scores under hypothetical currency-volatility scenarios, and bootstrapped indirect effects supported a statistical mediation pattern for all three biases between social-media reliance and investment decisions. A direct test of a Prospect Theory-inspired loss/gain framing difference comparing responses to loss-framed and gain-framed depreciation/appreciation scenarios did not find a statistically significant difference, suggesting that simple gain/loss framing alone may not be sufficient to produce the asymmetric risk response the theory predicts in this macro-financial context. Because the design is cross-sectional, the sample was drawn through purposive sampling of investors active in online forums, and currency volatility was operationalised through investors’ perceptions of hypothetical scenarios rather than objective exchange-rate data; the findings should be read as associations that describe this sample rather than as causal or population-representative claims. The study nonetheless offers an internally consistent, re-estimated account of how social-media reliance and financial literacy relate to specific behavioural biases in the Indian mid-cap segment, with implications for investor-education design that are stated as directions worth testing rather than as validated interventions.