Marcel Quint

Working Papers

[JMP] The Interaction of Memory Imperfections (with Johannes Maier)

Abstract: We investigate how two pervasive imperfections of human memory – motivated recall (remembering “good” rather than “bad” information) and similarity-based recall (remembering information similar to current information) – interact. We propose a theoretical model of recall in which these biases can reinforce or offset each other, depending on the cognitive effort invested in recall. With lower effort exerted, the two biases are more complementary, as people rely more on intuitive reasoning and thereby become more able to exploit similarity to self-servingly bias their recall. We test the model’s predictions in a laboratory experiment that varies both motivation and contextual similarity. Consistent with the model, we find that the two biases are complements on average, in particular when recall effort is low. The interaction shapes beliefs and behavior (such as trading and policy responses), so that simply “adding up” the biases’ isolated effects would lead to misguided inferences under their coexistence. We further show that imperfect memory is equally important in explaining distorted beliefs as non-Bayesian updating.

Counterfactual Thinking Among Retail Investors (with Lukas Mertes) 

Abstract: We demonstrate that individuals consider counterfactual outcomes when evaluating their investment decisions. Using individual investor trading data from the U.S., we show that investors are more likely to sell a stock that outperformed an alternative investment than a stock that underperformed the alternative. This pattern exists when considering the overall market as well as the stock's industry as alternative investment. The effect is distinct from (and can even subsume) the Disposition Effect, and cannot be explained by previous findings, investors who risk-adjust, or by stock characteristics. We replicate our findings out-of-sample using more recent German trading data. As an underlying mechanism, we provide evidence that investors' reference points shift across market phases, making the market a salient counterfactual during boom markets. Our results highlight that individuals evaluate their financial investments not in isolation, but relative to the investments they could have made instead, particularly during market upswings. 

Work in Progress

Climate Experiences and Financial Behavior

Abstract: I investigate whether personal experiences with climate catastrophes shape financial decision making. Using county level data on climate risk beliefs and catastrophe exposure records for the United States, I first show that experiencing a climate catastrophe substantially increases concern about future risks related to climate change, even years after the event occurred. To isolate the role of personal experience from that of information, I exploit within state variation in catastrophe exposure in a staggered Difference in Differences design. My findings show that personal experience has a significant and persistent effect on climate risk beliefs beyond informational channels alone. Second, I am analysing whether these belief shifts translate into financial decision making. I use policy level records from the National Flood Insurance Program (NFIP) to examine whether affected counties exhibit higher take up rates following a catastrophe. Using an instrumental variables strategy that instruments climate beliefs with catastrophe exposure (excluding flooding), I can isolate the belief channel from institutional responses that floods trigger directly. Preliminary results suggest that climate experiences not only shift what people believe but also how they make financial decisions. 

Motivated Reference Points in Financial Decisions (with Lukas Mertes)

Abstract: When evaluating their investment performance, investors can compare their outcomes to multiple potential reference points, such as alternative investments, purchase prices, market indices, or their own expectations. However, little is known about how investors select among these reference points. This paper proposes a model in which investors choose reference points ex-post to minimize regret, creating a trade-off between accuracy and regret avoidance. We aim to conduct an online experiment to test the predictions of our model, particularly whether investors shift their reference points towards less regrettable comparisons when investment performance falls short of expectations, despite the associated cost of holding distorted beliefs. This mechanism should lead investors to hold overly optimistic beliefs about  their own investments, ultimately affecting their trading behavior.