Dynamic Autoregressive Liquidity (DArLiQ)
Hafner, C. M.
Linton, O. B.
Cambridge Working Papers in Economics
Janeway Institute Working Paper Series
Faculty of Economics, University of Cambridge
MetadataShow full item record
Hafner, C. M., Linton, O. B., & Wang, L. (2022). Dynamic Autoregressive Liquidity (DArLiQ). https://doi.org/10.17863/CAM.82497
Motivated beliefs theory suggests the absorption of information may be biased, especially when it bears consequences for the ego. This paper finds empirical support for that hypothesis in the field, using longitudinal data on teenagers’ memories of mathematics report card grades and administrative data on actual grades. Students: i) make more errors in recalling lower grades; ii) update their academic self-confidence in association with recalled grades rather than actual grades; and iii) have more flattering memories of grades when the survey was administered with a longer delay. The first two results bolster recent research in demonstrating that patterns of motivated recall are robust to within-individual estimation. The last result extends the field literature in showing that a large part of the mechanism for motivated information absorption is memory loss over time. A structural model is used to represent memories as the outcome of a subconscious choice problem, disentangling competing motives to enhance self-confidence and respect reality. The estimated model indicates that the costs of memory distortions decrease as time passes after information transmission, and students with low self-confidence had a greatly diminished preference for inflating self-confidence via memory distortions
Nonparametric, Semiparametric, Splits, Structural Change
This record's DOI: https://doi.org/10.17863/CAM.82497
This record's URL: https://www.repository.cam.ac.uk/handle/1810/335058
All Rights Reserved
Licence URL: https://www.rioxx.net/licenses/all-rights-reserved/