Emotionally Rational Consumer's Demand: Empirical Analysis of Fashion Industry

Abstract

Consumers are not only rational, but also they have emotion. In other word consumers incur mental cost, or emotion, for their forgone decisions, and learning from past mistakes shows itself in consumer's decision in the form of counterfactual thinking. Counterfactual thinking if incorporated into utility function of consumer predicts choice better. This study develops and estimates a two-period model of consumer choice in markets of fashion goods. Fashion goods are distinct from other types of goods in that they have limited lifecycle and availability. In these markets, retailers use skimming strategy by exerting multiple markdowns during the season. In reaction to this retail strategy, emotionally rational consumer faces a trades between purchasing early in the season, taking a risk of high price regret, or delaying it to the second period, when the product may not be available anymore, taking a risk of inaction or unavailability regret. Our structural model not only allows for four main component of consumer's decision on fashion items, two rational and two emotional. The first rational component includes ownership utility, consumption utility, and unobserved utility, or demand shock. The second rational component includes price disutility. The first emotional component in our model includes high price regret, and the second one inaction or stock out regret. We first assume that consumer has rational expectation not only about her regret sensitivity, but also about availability, price and time of markdown. We further extend our model to account for consumer misconception, and counterpart of regret, or rejoice of purchasing under markdown. We first start modeling myopic and forward looking consumer decision separately, but then we build up the model to allow for both type of consumer segments. Our model allows heterogeneity of consumer response to both price and different types of regret. We build up our model to account for multiple categories, to test for different hypothesis on consumer's regret on cold and hot whether, and impact of gender and product type on regret and price sensitivity. Finally, we extend our model for multiple periods. We estimate our model using aggregate sales and inventory data from a specialty apparel retailer. [the results are subject to change] We find out that there is small segment of high price sensitive consumers who are more regretful to high price, but less regretful to stock out. We also find out that if we do not account for heterogeneity in consumer price sensitivity, and consumer regret estimates are biased. Finally, our results show that only accounting for supply side optimization of firm does not substitute heterogeneity in consumer response.