Emil Bjerre Jensen forsvarer sin ph.d.-afhandling

Emil Bjerre jensen forsvarer sin ph.d.-afhandling:"Essays on Spending, Mortgages and Preferences. Evidence from transaction data"


Emil Bjerre jensen


"Essays on Spending, Mortgages and Preferences. Evidence from transaction data". Det vil være muligt før forsvaret at rekvirere en kopi af afhandlingen ved henvendelse til Receptionen (26.0.20), Økonomisk Institut.

Tid og sted

13. december 2019 kl. 13:00, Økonomisk Intitut, Københavns Universitet, Øster Farimagsgade 5, 1353 København K, bygning 35, lokale CSS 35.3.12. Af hensyn til kandidaten lukkes dørene præcis.


Lektor Asger Lau Andersen, Økonomisk Institut, Københavns Universitet, Danmark (formand)
Lektor Kaveh Majlesi, Lund Universitet, Sverige
Professor Gisle Natvik, BI Norwegian Business School, Norge


This thesis consists of three self-contained chapters. In the first chapter, written together with Jeppe Druedahl and Søren Leth-Petersen, we study the spending responses of borrowers with adjustable rate mortgages (ARM) around the timing of an interest rate reset. We exploit that the bank sends a letter to ARM borrowers in advance where they inform about the expected change in mortgage payments upon the reset. We find that unconstrained households respond immediately, while liquidity constrained households respond around the time where cash-flow arrives, and rationalize the behavior within a buffer-stock model. 

In the second chapter, written together with Søren Leth-Petersen, we examine whether home owners, who refinance fixed rate mortgages, increase spending. We find that the overall spending effect is small. To understand this result, we consider two groups of refinancers. One group face a strong incentive to refinance to lock in a lower market rate, and are primarily found to increase loan repayments and saving. Another group of refinancers have less of an incentive to refinance, and we find that they, on the contrary, are more likely to cash-out refinance and increase spending. 

In the third chapter, I examine how household-specific preferences predict spending responses by structurally estimating preference parameters non-parametrically. Based on the structurally estimated preference parameters, I calibrate household-specific consumption functions to derive MPC estimates. By decomposing the variance of the MPC, I show that 76% of the variation in the MPC is driven by characteristics while 24% is due to temporary circumstances. In a simulation exercise, I demonstrate how household characteristics create a deadweight loss under fiscal polices due to significant differences in spending responses."